The best reason why to use open-source software. IMHO, that is.

Christian Klippel · 14th July 2009, 04:12 PM

Hello,

first of all, it's not my intention to bash commercial software companies here. While i'm referring to certain softwares, the same holds true for other, closed-source packages. Also, i'm not about communism/socialism in software, or anything like that. I'm simply trying to give, in my opinion, a strong argument for open source.

I try to be simple and to focus on a narrow example. Also, i'd like to hear about your point of view on these matters.

The question is: Who owns your data?

You may think "well, i own it, who else?", but is that really so?

Probably you have experienced the following situation already: a friend or business partner sends you a Word-document. Because he/she/the company uses always the latest and greatest versions, you got a file that can not be opened in your version, which is just the one before. OK, you may use the latest free Wordpad to open it, but you can not open it in your word-processor. If that happens more often, you are probably forced to upgrade your version as well, if you want to interchange documents.

So far not the worst situation, you say. However, your older version might be just fine for you since it does everything you want. It could also be the other way round. You send someone a document that you created with an old version, and the recipient has problems to open it. He might be able to open it, but the formatting is all wrong. Such scenarios are far from being unusual.

Now, take it a step further. All your documents, may they be for your private stuff, or, even worse, for your business, are created using a certain package. Of course, all these formats are rather undocumented. And they may have some really ugly quirks as well. Just look at the spec's for OfficeOpen XML from Microsoft for Excel, when it comes to date formats, numbers, formulas.

However, the vendor goes south. Sure, that might not happen soon with Microsoft. But it may happen. Or you don't want to invest more and more money to keep your softwares (and i mean only the applications) to the latest and greatest. You only update your OS. But some day, your old version doesn't run anymore on the new OS. So, you are again forced to upgrade your programs.

But, what is with the data? You can not simply take it and use it in a different software-package. Well, you can, to be true. But you have to expect troubles when doing so. You would need to invest lots of hours. Since the structure of the proprietary file formats is not really known to anyone else, you will have problems to read them with another package.

Also, you never know what is actually saved in the files, because you don't know the internals. For example, there have been many cases where old contents of a .DOC file have been recovered, just to reveal some compromising content. Imagine you make an offer to a customer. You use your standard letter, just change it and save the new version in an extra file. However, chances are that there is still content of the old, original letter in the new one. What if your customer finds out, through that way, that you make much higher prices for him than for your other customer? What if he finds out that some person he hates has relations to your company, just because it leaked through such a file?

With open-source, you can avoid these problems. In case some software ceases to exist, you still have the full spec's of the file format in which your documents are saved. Chances are very high that other open-source programs can handle these files just fine already, because the full implementation is open. And if it really goes bad, you still have the full spec's and an open reference implementation to maintain access to your data.

You can be sure that no compromising residues are stored in the files, because you can look at them and understand what is where. OK, maybe not you personally. But a lot of people can. And they would speak up if this was the case. Also, open-source programmers will surely avoid such issues right from the start.

Admitted, open-source is not perfect. But so is commercial software, maybe to some lesser extent, but still. Closed-source is black-box where you have no idea what goes on, and no chance to recover when things go south. With open-source, you can always access the information in your files, you can hire a programmer to customize the program exactly the way you want it.

In any case, with open-source, it is _you_ who owns your data. You will always have full access to it, and you can always be fully aware of what is going on. Can you say the same for closed-source black-box software?

In the private sector it might not have that much weight. Letters to mom and dad lost? Who cares, let's write new ones. Probably a PITA, and bad to have lost the old "memories", but so be it. Now, what in the case of a business? Do you really want to fully depend on some external company to have access to your very own data?

So, who owns your data?

Greetings,

Chris

ddt · 16th July 2009, 01:48 AM

Your argument is not so much a reason to use open-source software, but a reason to use open standards. The two often go hand-in-hand - and you can even argue that an open source product by definition uses an open standard (after all, the source code defines the file format

) - but that need not be. Hasn't Microsoft itself written an ODF plugin for MS Word?

laca · 16th July 2009, 03:33 AM

Originally Posted by ddt

Your argument is not so much a reason to use open-source software, but a reason to use open standards. The two often go hand-in-hand - and you can even argue that an open source product by definition uses an open standard (after all, the source code defines the file format

) - but that need not be. Hasn't Microsoft itself written an ODF plugin for MS Word?

While the open source software is open, that does not mean it is using proper standards, i.e. its formats may not be standardized at all, only open.

I agree though that a closed-source product does not necessarily use closed formats.

ddt · 16th July 2009, 03:52 AM

Originally Posted by laca

While the open source software is open, that does not mean it is using proper standards, i.e. its formats may not be standardized at all, only open.

You're absolutely right. To which I can counter: when does a format become a standard?

laca · 16th July 2009, 04:02 AM

Originally Posted by ddt

You're absolutely right. To which I can counter: when does a format become a standard?

When it is submitted to an appropriate body of standardization (e.g. ISO or ECMA) and it is accepted and given a standards number (e.g ODF is ISO/IEC 26300:2006).

Christian Klippel · 16th July 2009, 05:57 AM

Hello ddt,

Originally Posted by ddt

Your argument is not so much a reason to use open-source software, but a reason to use open standards.

well, yes and no i would say. Sure, having the format being an open standard is part of the "deal". However, standards are usually a big chunk of dead trees that in no way guarantee that you can really always handle the files that are created.

Look at the fuss around OOXML from Microsoft, for example. Thousands of pages, but still not really complete. In some places proprietary extensions kick in that are not defined there at all. Also there is the ever-present patent-threat hanging over OOXML. Having a sepc is one thing, being allowed to actually implement it is another.

However, that is a not problem with Microsoft per se, but more with a borked-up patent system that allows patents on software.

With open-source programs you also have an open, working implementation of the standard that you can refer to.

Maybe i didn't made that point clear enough in my OP.

Originally Posted by ddt

The two often go hand-in-hand - and you can even argue that an open source product by definition uses an open standard (after all, the source code defines the file format

) - but that need not be. Hasn't Microsoft itself written an ODF plugin for MS Word?

Yes, they did, and doesn't really work when it comes to interchange the files across different programs. Admitted, that is partly because the current accepted ODF spec's (there are new ones coming up, but they haven't gone through the ISO/ECMA/whatever process yet) lack detail in certain areas, for example how cell values, formulas and such are stored in spreadsheet files. On the other hand, they implemented them in their own way, instead of looking how existing softwares do that, so they could stay compatible.

And IIRC the plugin was only financed by Microsoft, but actually written by an external company. So it's not really them to blame.

Greetings,

Chris

Wudang · 16th July 2009, 03:30 PM

There's also standards and implemenations of standards. Consider the SQL standards - they do not always guarantee that the same SQL query ot 2 SQL 92 standard RDBMs will return the same result set in the same order.
Neither I suspect would an open standard for say documents guarantee that 2 compliant word processors would produce identically formatted results.

Other than that - **** yeah for open standards.

MetalPig · 17th July 2009, 02:14 AM

Originally Posted by Wudang

Consider the SQL standards - they do not always guarantee that the same SQL query ot 2 SQL 92 standard RDBMs will return the same result set in the same order.

Even if you specify the order in the query? I find that hard to believe.

SezMe · 17th July 2009, 02:22 AM

Originally Posted by Christian Klippel

However, that is a not problem with Microsoft per se, but more with a borked-up patent system that allows patents on software.

What does "borked-up" mean?

daSkeptic · 17th July 2009, 03:03 AM

Originally Posted by MetalPig

Even if you specify the order in the query? I find that hard to believe.

Result-set, not column-set.

laca · 17th July 2009, 03:44 AM

Originally Posted by SezMe

What does "borked-up" mean?

http://www.urbandictionary.com/define.php?term=borked

MetalPig · 17th July 2009, 05:33 AM

Originally Posted by daSkeptic

Result-set, not column-set.

I know. I was referring to the use of an 'order by'-clause.

Wudang · 17th July 2009, 06:56 AM

Are NULLs less than or more than non-NULLs?

It wasn't a very good example really was it?

MetalPig · 17th July 2009, 09:05 AM

Originally Posted by Wudang

Are NULLs less than or more than non-NULLs?

My answer would be: neither.
What do the SQL standards say about it?

Wudang · 17th July 2009, 09:50 AM

It's implementation defined so for queries where NULLs are preserved like a LEFT OUTER JOIN the order may be different for different vendors. Of course this is because the standard didn't specify the precedence of NULLs so as I said it's not as good an anlogy as it seemed to my addled brain at the time.

shadron · 17th July 2009, 01:06 PM

Originally Posted by MetalPig

Even if you specify the order in the query? I find that hard to believe.

In fact, SQL doesn't at all specify the order of rows returned, it is undefined, in the absence of an ORDER BY clause. I got that drilled into me at one point when we segmented the execution of a 5 million record return set. We experimented by running a segment of output, doing a checkpoint and then running the next segment. When the checkpoint was returned to, the order of the returned rows in the second segment changed significantly. No one ever said you couldn't run from a checkpoint and expect the same results; the programmer just assumed it.

The moral was always to specify an ORDER BY clause, or else don't checkpoint. ORDER BY significantly slows down results return, and delays it until the entire results set is known. Hurts a lot.

ddt · 17th July 2009, 01:11 PM

Originally Posted by daSkeptic

Result-set, not column-set.

Yes, but you can specify order of rows, like this:

SELECT name, age OF person ORDER BY age DESCENDING

Seems clear-cut to me then how to order them.

ETA: I should have refreshed that page before replying.

Improbable Joe · 18th July 2009, 08:27 PM

Originally Posted by Christian Klippel

However, that is a not problem with Microsoft per se, but more with a borked-up patent system that allows patents on software.

"Allows"?!?! Shouldn't software developers get paid, or should they exist as your slave, for your benefit? I'm not a crazy free market capitalist or anything(I tend to mock them) but eliminating the patents on software sounds like an ethical disaster, and the end of all sorts of technological development.

Maybe that's what you want: a wasteland where you never pay for anything, and all you get is almost-professional software that only works for almost-professional users. The reality is that Microsoft deserves every penny it gets, because it has designed software that works relatively well for pretty much everyone.

shadron · 18th July 2009, 09:09 PM

The problem isn't with the fact that software developers and their ~~slave drivers~~ - uhhh, employers shouldn't get paid for the former's work, but rather that the patent system is a blunt instrument for doing that. Rather than having something similar to a copyright law that keeps one's code under the ownership of the writer or his assignee, the patent system is used to freeze the ability for others to use features that could be implemented in many different ways. For example, Microsoft has patents on a number of Windows features which make it impossible for anyone else to duplicate the Windows "look and feel" without Microsoft's license, or the long-running battle over the use of gif image format, which kept that format from becomeing a lot more influential than it really was.

No need to take off after Christian for Microsoft's sake, they've amply shown thay can take care of themselves; in fact, if you read carefully, Christian doesn't castigate M$ but rather the American court's upholding patent law for software ideas.

Modified · 18th July 2009, 09:13 PM

Originally Posted by Improbable Joe

"Allows"?!?! Shouldn't software developers get paid, or should they exist as your slave, for your benefit? I'm not a crazy free market capitalist or anything(I tend to mock them) but eliminating the patents on software sounds like an ethical disaster, and the end of all sorts of technological development.

Maybe that's what you want: a wasteland where you never pay for anything, and all you get is almost-professional software that only works for almost-professional users. The reality is that Microsoft deserves every penny it gets, because it has designed software that works relatively well for pretty much everyone.

Software will be covered by copyright, either way. I can see that maybe a truly novel and as yet unpublished algorithm should be patentable, but that covers approximately zero software patents. Microsoft just applies for patents for everything they do, even if it is copied directly from someone else (http://blogs.zdnet.com/Burnette/?p=245 for example). Many of their patent applications are vague or unintelligible. The "inventors" listed on the patent application probably have never even seen it in most cases.

The problem is, any significant software that anyone writes is likely to infringe on several existing software patents, even if the authors believe that every idea and algorithm they are using is unique. For the most part, these patents will never be enforced, and large companies use them mainly for lawsuit defense.

Improbable Joe · 18th July 2009, 09:23 PM

Originally Posted by Modified

Software will be covered by copyright, either way. I can see that maybe a truly novel and as yet unpublished algorithm should be patentable, but that covers approximately zero software patents. Microsoft just applies for patents for everything they do, even if it is copied directly from someone else (http://blogs.zdnet.com/Burnette/?p=245 for example). Many of their patent applications are vague or unintelligible. The "inventors" listed on the patent application probably have never even seen it in most cases.

The problem is, any significant software that anyone writes is likely to infringe on several existing software patents, even if the authors believe that every idea and algorithm they are using is unique. For the most part, these patents will never be enforced, and large companies use them mainly for lawsuit defense.

It sounds like sour grapes to me. There's nothing that is so wrong with Microsoft that the answer is an end to copyright on software, as has been both explicitly and implicitly suggested in this thread.

Laws are never perfect, but the idea that the law should be abandoned because people are whining about Microsoft? Not so much... and I'm not even a huge fan of the company. I should correct an earlier statement: Microsoft deserves MOST of the money they get... they can be a giant PITA, but do a good job overall. They can be proud of the job they've done, compared to some companies, even though they have huge room for improvement.

My position is that when dealing with these sort of issues, the system isn't the problem, as much as specific vendors in very narrow cases. you need to consider going about it with a laser scalpel, instead of using a chainsaw to cut off entire legal principals.

Modified · 18th July 2009, 10:42 PM

Originally Posted by Improbable Joe

It sounds like sour grapes to me.

I'm not sure what you mean. I haven't applied for any software patents, and even if I had I would think they were crap.

Quote:

There's nothing that is so wrong with Microsoft that the answer is an end to copyright on software, as has been both explicitly and implicitly suggested in this thread.

I don't want software patents scrapped to punish Microsoft, I want them scrapped because they are mostly junk. Microsoft would probably be better off without them, since they use them almost exclusively for defense (not to prevent others from using their ideas, but to prevent themselves from being sued for using others' ideas).

Quote:

My position is that when dealing with these sort of issues, the system isn't the problem, as much as specific vendors in very narrow cases.

Then you don't know anything about it. As I said, any significant software that is created is likely to violate multiple patents. Determining which patents may be violated would cost many times the software development cost. Creating many common classes of software without violating patents is impossible. Imagine there were thousands of laws prohibiting common activities that you were unaware of and were rarely or never enforced, but for which you could potentially be punished and which in many cases you could not avoid violating.

laca · 19th July 2009, 05:53 AM

Originally Posted by Improbable Joe

"Allows"?!?! Shouldn't software developers get paid, or should they exist as your slave, for your benefit? I'm not a crazy free market capitalist or anything(I tend to mock them) but eliminating the patents on software sounds like an ethical disaster, and the end of all sorts of technological development.

You are confusing patents with copyright. What if Van Gogh patented a color so noone else could use it? Mind you, not the batch he made, but the color itself. Now that is what ends "all sorts of technological development". In that kind of world there can never be a painter who sees one of Van Gogh's paintings, loves that color and tries to reproduce it.

Originally Posted by Improbable Joe

Maybe that's what you want: a wasteland where you never pay for anything, and all you get is almost-professional software that only works for almost-professional users.

Strawman. Nobody said that.

Originally Posted by Improbable Joe

The reality is that Microsoft deserves every penny it gets, because it has designed software that works relatively well for pretty much everyone.

Does Microsoft also deserve to be the target of patent trolls?

laca · 19th July 2009, 05:59 AM

Originally Posted by Modified

Software will be covered by copyright, either way. I can see that maybe a truly novel and as yet unpublished algorithm should be patentable, but that covers approximately zero software patents.

Hmmm... How about patenting mathematical theorems? To me, that is not so much different from patenting algorithms.

Originally Posted by Modified

Microsoft just applies for patents for everything they do, even if it is copied directly from someone else (http://blogs.zdnet.com/Burnette/?p=245 for example). Many of their patent applications are vague or unintelligible. The "inventors" listed on the patent application probably have never even seen it in most cases.

Maybe. But not only Microsoft does it. Almost everyone does it. There are companies that specialize in patents. That is they register patents in the hope to extort some money from them in the future.

Originally Posted by Modified

The problem is, any significant software that anyone writes is likely to infringe on several existing software patents, even if the authors believe that every idea and algorithm they are using is unique. For the most part, these patents will never be enforced, and large companies use them mainly for lawsuit defense.

Agreed 100%.

laca · 19th July 2009, 06:01 AM

Originally Posted by Improbable Joe

It sounds like sour grapes to me. There's nothing that is so wrong with Microsoft that the answer is an end to copyright on software, as has been both explicitly and implicitly suggested in this thread.

Would you mind pointing out who suggested to end copyright on software in this thread? I can't find it and I would have some stuff to say about that as well...

Christian Klippel · 19th July 2009, 07:49 AM

Originally Posted by Improbable Joe

"Allows"?!?! Shouldn't software developers get paid, or should they exist as your slave, for your benefit? I'm not a crazy free market capitalist or anything(I tend to mock them) but eliminating the patents on software sounds like an ethical disaster, and the end of all sorts of technological development.

Maybe that's what you want: a wasteland where you never pay for anything, and all you get is almost-professional software that only works for almost-professional users. The reality is that Microsoft deserves every penny it gets, because it has designed software that works relatively well for pretty much everyone.

I think that you have no grasp about the issues at hand. Software is covered by copyright protection already, and that in fact is the proper way to protect it. Maybe you are not aware of the difference between a patent and between a copyright.

And before you go of and try to tell me that i want to smear or punish Microsoft, or that i want every programmer to not be paid for her/his work, go back and read what i wrote.

The problem with software patents is that software is not a physical thing. It is not an object that has a dedicated function and needs a special way to be produced. Software is much more like literature. You use common words, and lots of them, to create a program. Pretty much the same way that you use a lot of words to write a book.

There are many, many things in software that simply can not be done in a different way. Well, maybe they could, but that would result in an really awkward pile of crappy code.

Did you know that the progress bar, that virtually every user interface uses, is actually patented? Look here.

Take a look here for more absurd software patents.

Again, software is already protected by copyright. And that is a good thing. The nastiness starts when software is patented, because that is simply a stupid thing to do. Nowdays it is virtually impossible to write a program that goes beyond the "Hello world!" example without infringing some stupid patent.

This blocks innovation, hinders competition and is a serious threat to the economy.

Greetings,

Chris

Christian Klippel · 19th July 2009, 08:02 AM

Originally Posted by Improbable Joe

It sounds like sour grapes to me. There's nothing that is so wrong with Microsoft that the answer is an end to copyright on software, as has been both explicitly and implicitly suggested in this thread.

No, that is not what has been suggested. Read my other post. And while you are at it, read the whole thread again.

Originally Posted by Improbable Joe

Laws are never perfect, but the idea that the law should be abandoned because people are whining about Microsoft? Not so much... and I'm not even a huge fan of the company. I should correct an earlier statement: Microsoft deserves MOST of the money they get... they can be a giant PITA, but do a good job overall. They can be proud of the job they've done, compared to some companies, even though they have huge room for improvement.

As i have said very clearly, this is not about Microsoft alone. It is about the general state of things, and i picked Microsoft simply as a vehicle to demonstrate my point. I explicitly said that this is not a problem of Microsoft in general.

Originally Posted by Improbable Joe

My position is that when dealing with these sort of issues, the system isn't the problem, as much as specific vendors in very narrow cases. you need to consider going about it with a laser scalpel, instead of using a chainsaw to cut off entire legal principals.

You are wrong. When a system allows for such things like software patents, that really threaten the market, then the problem is indeed the system, and not the companies. Companies only do what the laws allow them to do. If the laws allow them to issue software patents, then they will do. And of course they will then go on and use this patents to threaten or even extinguish competition. It is not the companies fault, the fault is with the system.

In any case, that hasn't much to do with my OP anyways. While patents on certain file formats play a role in that (see the GIF patent, for example), the basic problem is that with many commercial software packages you have no real control over your data. You are forced to stick to the programs without having a decent way out. If everything goes wrong you are left with a bunch of binary files that you can not read anymore. If such a thing happens to a companies files, i guess that would be a disastrous event.

Sure, one can sit down and try to export the files from an old version of a given program, and then import them into a different program. But that task is extremely time consuming and error prone. You surely will loose the formatting, font's, etc. So you would have to re-work them all. Exporting and re-importing databases is an even more complex task to do.

IMHO, the only way to avoid such situations is to use open standards that are implemented as open source, because only then you have full access to the data, because of the open-source implementation you have some kind of a reference: namely the way the files were created/written and are read back in the first place.

Greetings,

Chris

Christian Klippel · 19th July 2009, 08:13 AM

BTW, the software-patents problem is not genuine to the USA anymore. Here in Europe we have the very same problems by now, albeit maybe not that extreme.

To make the problem a bit more clearer to people that are not into classical software development, take a classical webshop.

Take a look at this page to see what stupid patents affect a simple webshop.

Then go here to read about how such patents can become weapons.

Now, if you have taken a look at the webshop example, take a moment to think about it. How could the used examples be implemented differently? Is there any way to implement them in a different way at all?

These are patents on abstract and very trivial things. Every programmer or web-designer with more than one brain cell will come to the very same solutions for the problems at hand.

Greetings,

Chris

Wudang · 19th July 2009, 08:57 AM

To be fair to Microsoft, when they splt with IBM over OS/2 and said to IBM "Windows is our ball and you can't play, ner ner". IBM replied with its patent inventory and said "well for starters you can stop using cursors becuase we own that patent". As well as this warehouse full.
In IBM Hursley we were encouraged to submit out ideas for patent filing but it was quite a process of technical evaluation etc. I had to pull mine when I found someone beat me to it.

Modified · 19th July 2009, 10:47 AM

Originally Posted by laca

Maybe. But not only Microsoft does it. Almost everyone does it. There are companies that specialize in patents. That is they register patents in the hope to extort some money from them in the future.

Apart from the patent trolls, most small software companies do not do this.

Microsoft does not seem to worry about prior art even when they are well aware of it. Other large companies may be as guilty, but I not seen such obvious cases that got any publicity.

Modified · 19th July 2009, 10:57 AM

Originally Posted by laca

Hmmm... How about patenting mathematical theorems? To me, that is not so much different from patenting algorithms.

As I indicated, I have difficulty with the idea. For a truly novel algorithm, something that took significant time and insight to develop but which is difficult to hide from reverse-engineers of the code, it would be nice to have some intellectual property protection for some length of time. Determining that an algorithm is truly novel, significant, and non-obvious to the degree required would be very difficult.

Modified · 19th July 2009, 11:05 AM

Originally Posted by Christian Klippel

Did you know that the progress bar, that virtually every user interface uses, is actually patented? Look here.

At least that one is well written.

laca · 19th July 2009, 01:27 PM

Originally Posted by Modified

As I indicated, I have difficulty with the idea. For a truly novel algorithm, something that took significant time and insight to develop but which is difficult to hide from reverse-engineers of the code, it would be nice to have some intellectual property protection for some length of time.

Again, how is this different from a new mathematical theorem?

Originally Posted by Modified

Determining that an algorithm is truly novel, significant, and non-obvious to the degree required would be very difficult.

A truly novel algorithm that took significant time and insight to develop might be recognized by requiring about the same time and insight to reverse engineer as to develop.

Modified · 19th July 2009, 02:54 PM

Originally Posted by laca

Again, how is this different from a new mathematical theorem?

Because they are two different things? I think you mean how is that different from an algorithm without an intended use in software. If an algorithm is used for some other purpose, then it could possibly be maintained as a trade secret, otherwise, if it is useful in some way but must be exposed in the process, then it should be treated no differently from a software algorithm.

Quote:

A truly novel algorithm that took significant time and insight to develop might be recognized by requiring about the same time and insight to reverse engineer as to develop.

I was talking about reverse engineering the code to discover the algorithm.

laca · 19th July 2009, 03:58 PM

Originally Posted by Modified

Because they are two different things?

Both are thought processes expressed in some way. And thought should not be patentable. IMHO.

Originally Posted by Modified

I think you mean how is that different from an algorithm without an intended use in software. If an algorithm is used for some other purpose, then it could possibly be maintained as a trade secret, otherwise, if it is useful in some way but must be exposed in the process, then it should be treated no differently from a software algorithm.

Are you suggesting that mathematical theorems have no use? What I'm trying to say is that an algorithm and a theorem are both tools. Neither is an end product. Well, most of the time.

Originally Posted by Modified

I was talking about reverse engineering the code to discover the algorithm.

Well so was I. Been there, done that. It is hard enough to reverse engineer a stupid copy protection scheme, not to mention a complex algorithm.

Modified · 20th July 2009, 12:28 AM

Originally Posted by laca

Both are thought processes expressed in some way. And thought should not be patentable. IMHO.

A theorem is not a thought process in the usual sense. Perhaps you mean a proof of a theorem?

Quote:

Are you suggesting that mathematical theorems have no use?

No. I thought you were attempting to make a distinction between software algorithms and mathematics rather than between theorems and algorithms.

Quote:

Well so was I. Been there, done that. It is hard enough to reverse engineer a stupid copy protection scheme, not to mention a complex algorithm.

Complex copy protection schemes are broken regularly. If someone were to develop, for example, a sorting algorithm that is on average much faster than any existing one for certain common types of data, it could be well worth the effort of reverse engineering. Does that development work deserve less protection than the development of a type of hammer that is more effective for certain uses? Does protection of one encourage innovation but the other not? I don't think so, but I do think that protection is more problematic for algorithms, perhaps so much so that overall we are better off without it.

laca · 20th July 2009, 12:58 AM

Originally Posted by Modified

A theorem is not a thought process in the usual sense. Perhaps you mean a proof of a theorem?

Yes, it is actually the end result of a thought process. Anyway, both are similar in that they from a starting point through some transformations arrive at something of "value" so to say.

Originally Posted by Modified

Complex copy protection schemes are broken regularly.

Well yes, but algorithms not so often. Not even simpler ones.

Originally Posted by Modified

If someone were to develop, for example, a sorting algorithm that is on average much faster than any existing one for certain common types of data, it could be well worth the effort of reverse engineering. Does that development work deserve less protection than the development of a type of hammer that is more effective for certain uses?

No, it doesn't. However, what needs protected is the implementation, and not the idea. If the guy who developed the algorithm registers a patent for "a way to sort quickly <a specific type of data>" that effectively kills off innovation. What if someone else gets to the same end result by other means independently? Don't you think that kind of thing needs to be at least enabled if not encouraged?

Originally Posted by Modified

Does protection of one encourage innovation but the other not? I don't think so, but I do think that protection is more problematic for algorithms, perhaps so much so that overall we are better off without it.

Yes, it is very problematic for algorithms. Not to mention what the free flow of ideas means to humankind.

Almo · 20th July 2009, 01:48 PM

This is why I never send Word docs around. Seems pretty simple to me. Only use the right tool for the right job. The number of times I see people sending a Word doc when plain text would do fine is surprising.

ETA: But you're right about data ownership. That does become an issue if you're not careful.

Modified · 21st July 2009, 01:13 AM

Originally Posted by laca

Well yes, but algorithms not so often. Not even simpler ones.

Many of those copy protection schemes are algorithms or involve algorithms. I work in the reverse engineering / reengineering field (as a tool developer). Protecting algorithms is difficult.

Quote:

If the guy who developed the algorithm registers a patent for "a way to sort quickly <a specific type of data>" that effectively kills off innovation. What if someone else gets to the same end result by other means independently? Don't you think that kind of thing needs to be at least enabled if not encouraged?

The same goes for any patent though. Patents encourage innovation by rewarding the initial development and requiring disclosure.

laca · 21st July 2009, 06:20 AM

Originally Posted by Modified

Many of those copy protection schemes are algorithms or involve algorithms. I work in the reverse engineering / reengineering field (as a tool developer). Protecting algorithms is difficult.

What I'm suggesting is that a sufficiently novel andcomplex algorithm wouldn't need much (if any) explicit protection. Its sheer novelness and complexity should suffice.

Originally Posted by Modified

The same goes for any patent though. Patents encourage innovation by rewarding the initial development and requiring disclosure.

I'm not so sure. IMO patents should be used for compensating for the huge amounts of money that go into research. And I'm talking about millions. That is why the patent lasts for about 20 years. To make sure the inventor has enough time to recover the cost of research and more.

Now for the realm of algorithms the picture is quite different. Reasearch cost is almost nothing compared to that of a physical entity. Prototypes cost in the vicinity of zero. And that is why patents for software should not exist. Just like for ideas. The very least they should last for a very short amount of time, like months or maybe a few years. Tops.

14th July 2009, 04:12 PM	#1
Christian Klippel Muse Join Date: Feb 2008 Location: Ruhr Area in Germany Posts: 653	The best reason why to use open-source software. IMHO, that is. Hello, first of all, it's not my intention to bash commercial software companies here. While i'm referring to certain softwares, the same holds true for other, closed-source packages. Also, i'm not about communism/socialism in software, or anything like that. I'm simply trying to give, in my opinion, a strong argument for open source. I try to be simple and to focus on a narrow example. Also, i'd like to hear about your point of view on these matters. The question is: Who owns your data? You may think "well, i own it, who else?", but is that really so? Probably you have experienced the following situation already: a friend or business partner sends you a Word-document. Because he/she/the company uses always the latest and greatest versions, you got a file that can not be opened in your version, which is just the one before. OK, you may use the latest free Wordpad to open it, but you can not open it in your word-processor. If that happens more often, you are probably forced to upgrade your version as well, if you want to interchange documents. So far not the worst situation, you say. However, your older version might be just fine for you since it does everything you want. It could also be the other way round. You send someone a document that you created with an old version, and the recipient has problems to open it. He might be able to open it, but the formatting is all wrong. Such scenarios are far from being unusual. Now, take it a step further. All your documents, may they be for your private stuff, or, even worse, for your business, are created using a certain package. Of course, all these formats are rather undocumented. And they may have some really ugly quirks as well. Just look at the spec's for OfficeOpen XML from Microsoft for Excel, when it comes to date formats, numbers, formulas. However, the vendor goes south. Sure, that might not happen soon with Microsoft. But it may happen. Or you don't want to invest more and more money to keep your softwares (and i mean only the applications) to the latest and greatest. You only update your OS. But some day, your old version doesn't run anymore on the new OS. So, you are again forced to upgrade your programs. But, what is with the data? You can not simply take it and use it in a different software-package. Well, you can, to be true. But you have to expect troubles when doing so. You would need to invest lots of hours. Since the structure of the proprietary file formats is not really known to anyone else, you will have problems to read them with another package. Also, you never know what is actually saved in the files, because you don't know the internals. For example, there have been many cases where old contents of a .DOC file have been recovered, just to reveal some compromising content. Imagine you make an offer to a customer. You use your standard letter, just change it and save the new version in an extra file. However, chances are that there is still content of the old, original letter in the new one. What if your customer finds out, through that way, that you make much higher prices for him than for your other customer? What if he finds out that some person he hates has relations to your company, just because it leaked through such a file? With open-source, you can avoid these problems. In case some software ceases to exist, you still have the full spec's of the file format in which your documents are saved. Chances are very high that other open-source programs can handle these files just fine already, because the full implementation is open. And if it really goes bad, you still have the full spec's and an open reference implementation to maintain access to your data. You can be sure that no compromising residues are stored in the files, because you can look at them and understand what is where. OK, maybe not you personally. But a lot of people can. And they would speak up if this was the case. Also, open-source programmers will surely avoid such issues right from the start. Admitted, open-source is not perfect. But so is commercial software, maybe to some lesser extent, but still. Closed-source is black-box where you have no idea what goes on, and no chance to recover when things go south. With open-source, you can always access the information in your files, you can hire a programmer to customize the program exactly the way you want it. In any case, with open-source, it is _you_ who owns your data. You will always have full access to it, and you can always be fully aware of what is going on. Can you say the same for closed-source black-box software? In the private sector it might not have that much weight. Letters to mom and dad lost? Who cares, let's write new ones. Probably a PITA, and bad to have lost the old "memories", but so be it. Now, what in the case of a business? Do you really want to fully depend on some external company to have access to your very own data? So, who owns your data? Greetings, Chris
	__________________ Humber-physics 101: The treadmill has no ground equivalent. This means that the belt is not the road, but the Earth. ... That means the belt is also a privileged and unique perspective. If not then the treadmill collapses to the real world equivalent of a real treadmill, with different objects at different velocities in the same frame. Either way, no motion.

16th July 2009, 01:48 AM	#2
ddt Mafia Penguin Join Date: Dec 2007 Location: Netherlands Posts: 4,084	Your argument is not so much a reason to use open-source software, but a reason to use open standards. The two often go hand-in-hand - and you can even argue that an open source product by definition uses an open standard (after all, the source code defines the file format ) - but that need not be. Hasn't Microsoft itself written an ODF plugin for MS Word?
	__________________ Proud member of the Solipsistic Autosycophant's Group I think I see a signature - nathan

16th July 2009, 03:33 AM	#3
laca Muse Join Date: May 2009 Posts: 739	Originally Posted by ddt Your argument is not so much a reason to use open-source software, but a reason to use open standards. The two often go hand-in-hand - and you can even argue that an open source product by definition uses an open standard (after all, the source code defines the file format ) - but that need not be. Hasn't Microsoft itself written an ODF plugin for MS Word? While the open source software is open, that does not mean it is using proper standards, i.e. its formats may not be standardized at all, only open. I agree though that a closed-source product does not necessarily use closed formats.
laca Muse Join Date: May 2009 Posts: 739

16th July 2009, 03:52 AM	#4
ddt Mafia Penguin Join Date: Dec 2007 Location: Netherlands Posts: 4,084	Originally Posted by laca While the open source software is open, that does not mean it is using proper standards, i.e. its formats may not be standardized at all, only open. You're absolutely right. To which I can counter: when does a format become a standard?
	__________________ Proud member of the Solipsistic Autosycophant's Group I think I see a signature - nathan

16th July 2009, 04:02 AM	#5
laca Muse Join Date: May 2009 Posts: 739	Originally Posted by ddt You're absolutely right. To which I can counter: when does a format become a standard? When it is submitted to an appropriate body of standardization (e.g. ISO or ECMA) and it is accepted and given a standards number (e.g ODF is ISO/IEC 26300:2006).
laca Muse Join Date: May 2009 Posts: 739

16th July 2009, 05:57 AM	#6
Christian Klippel Muse Join Date: Feb 2008 Location: Ruhr Area in Germany Posts: 653	Hello ddt, Originally Posted by ddt Your argument is not so much a reason to use open-source software, but a reason to use open standards. well, yes and no i would say. Sure, having the format being an open standard is part of the "deal". However, standards are usually a big chunk of dead trees that in no way guarantee that you can really always handle the files that are created. Look at the fuss around OOXML from Microsoft, for example. Thousands of pages, but still not really complete. In some places proprietary extensions kick in that are not defined there at all. Also there is the ever-present patent-threat hanging over OOXML. Having a sepc is one thing, being allowed to actually implement it is another. However, that is a not problem with Microsoft per se, but more with a borked-up patent system that allows patents on software. With open-source programs you also have an open, working implementation of the standard that you can refer to. Maybe i didn't made that point clear enough in my OP. Originally Posted by ddt The two often go hand-in-hand - and you can even argue that an open source product by definition uses an open standard (after all, the source code defines the file format ) - but that need not be. Hasn't Microsoft itself written an ODF plugin for MS Word? Yes, they did, and doesn't really work when it comes to interchange the files across different programs. Admitted, that is partly because the current accepted ODF spec's (there are new ones coming up, but they haven't gone through the ISO/ECMA/whatever process yet) lack detail in certain areas, for example how cell values, formulas and such are stored in spreadsheet files. On the other hand, they implemented them in their own way, instead of looking how existing softwares do that, so they could stay compatible. And IIRC the plugin was only financed by Microsoft, but actually written by an external company. So it's not really them to blame. Greetings, Chris
	__________________ Humber-physics 101: The treadmill has no ground equivalent. This means that the belt is not the road, but the Earth. ... That means the belt is also a privileged and unique perspective. If not then the treadmill collapses to the real world equivalent of a real treadmill, with different objects at different velocities in the same frame. Either way, no motion. Last edited by Christian Klippel; 16th July 2009 at 07:20 AM. Reason: corrected bad quoting

16th July 2009, 03:30 PM	#7
Wudang BOFH Join Date: Jun 2003 Location: Sheffield Posts: 4,412	There's also standards and implemenations of standards. Consider the SQL standards - they do not always guarantee that the same SQL query ot 2 SQL 92 standard RDBMs will return the same result set in the same order. Neither I suspect would an open standard for say documents guarantee that 2 compliant word processors would produce identically formatted results. Other than that - **** yeah for open standards.
	__________________ Aphorism: Subjects most likely to be declared inappropriate for humor are the ones most in need of it.

17th July 2009, 02:14 AM	#8
MetalPig Muse Join Date: Jan 2006 Posts: 979	Originally Posted by Wudang Consider the SQL standards - they do not always guarantee that the same SQL query ot 2 SQL 92 standard RDBMs will return the same result set in the same order. Even if you specify the order in the query? I find that hard to believe.
MetalPig Muse Join Date: Jan 2006 Posts: 979	__________________ When I met you my eyes hurt That is how beautiful you are

17th July 2009, 02:22 AM	#9
SezMe Philosopher Join Date: Dec 2003 Location: Santa Barbara, CA Posts: 9,129	Originally Posted by Christian Klippel However, that is a not problem with Microsoft per se, but more with a borked-up patent system that allows patents on software. What does "borked-up" mean?

17th July 2009, 03:03 AM	#10
daSkeptic Muse Join Date: Aug 2008 Location: Bellingham, WA ... under the cheese counter at Haggen Posts: 712	Originally Posted by MetalPig Even if you specify the order in the query? I find that hard to believe. Result-set, not column-set.
	__________________ crud: noun - The SI unit of incredibility, equal to the movement of one-litre of hot air through a distance of one-metre in one-second. kudo: noun - The SI unit of credibility, reciprocal of the crud.

17th July 2009, 03:44 AM	#11
laca Muse Join Date: May 2009 Posts: 739	Originally Posted by SezMe What does "borked-up" mean? http://www.urbandictionary.com/define.php?term=borked
laca Muse Join Date: May 2009 Posts: 739

17th July 2009, 05:33 AM	#12
MetalPig Muse Join Date: Jan 2006 Posts: 979	Originally Posted by daSkeptic Result-set, not column-set. I know. I was referring to the use of an 'order by'-clause.
MetalPig Muse Join Date: Jan 2006 Posts: 979	__________________ When I met you my eyes hurt That is how beautiful you are

17th July 2009, 06:56 AM	#13
Wudang BOFH Join Date: Jun 2003 Location: Sheffield Posts: 4,412	Are NULLs less than or more than non-NULLs? It wasn't a very good example really was it?
	__________________ Aphorism: Subjects most likely to be declared inappropriate for humor are the ones most in need of it. Last edited by Wudang; 17th July 2009 at 08:02 AM. Reason: I need more sleep

17th July 2009, 09:05 AM	#14
MetalPig Muse Join Date: Jan 2006 Posts: 979	Originally Posted by Wudang Are NULLs less than or more than non-NULLs? My answer would be: neither. What do the SQL standards say about it?
MetalPig Muse Join Date: Jan 2006 Posts: 979	__________________ When I met you my eyes hurt That is how beautiful you are

17th July 2009, 09:50 AM	#15
Wudang BOFH Join Date: Jun 2003 Location: Sheffield Posts: 4,412	It's implementation defined so for queries where NULLs are preserved like a LEFT OUTER JOIN the order may be different for different vendors. Of course this is because the standard didn't specify the precedence of NULLs so as I said it's not as good an anlogy as it seemed to my addled brain at the time.
	__________________ Aphorism: Subjects most likely to be declared inappropriate for humor are the ones most in need of it.

17th July 2009, 01:06 PM	#16
shadron Illuminator Join Date: Sep 2005 Location: Colorado Posts: 3,856	Originally Posted by MetalPig Even if you specify the order in the query? I find that hard to believe. In fact, SQL doesn't at all specify the order of rows returned, it is undefined, in the absence of an ORDER BY clause. I got that drilled into me at one point when we segmented the execution of a 5 million record return set. We experimented by running a segment of output, doing a checkpoint and then running the next segment. When the checkpoint was returned to, the order of the returned rows in the second segment changed significantly. No one ever said you couldn't run from a checkpoint and expect the same results; the programmer just assumed it. The moral was always to specify an ORDER BY clause, or else don't checkpoint. ORDER BY significantly slows down results return, and delays it until the entire results set is known. Hurts a lot.
	Last edited by shadron; 17th July 2009 at 01:10 PM.

17th July 2009, 01:11 PM	#17
ddt Mafia Penguin Join Date: Dec 2007 Location: Netherlands Posts: 4,084	Originally Posted by daSkeptic Result-set, not column-set. Yes, but you can specify order of rows, like this: SELECT name, age OF person ORDER BY age DESCENDING Seems clear-cut to me then how to order them. ETA: I should have refreshed that page before replying.
	__________________ Proud member of the Solipsistic Autosycophant's Group I think I see a signature - nathan Last edited by ddt; 17th July 2009 at 01:13 PM.

18th July 2009, 08:27 PM	#18
Improbable Joe Thinker Join Date: Feb 2009 Posts: 147	Originally Posted by Christian Klippel However, that is a not problem with Microsoft per se, but more with a borked-up patent system that allows patents on software. "Allows"?!?! Shouldn't software developers get paid, or should they exist as your slave, for your benefit? I'm not a crazy free market capitalist or anything(I tend to mock them) but eliminating the patents on software sounds like an ethical disaster, and the end of all sorts of technological development. Maybe that's what you want: a wasteland where you never pay for anything, and all you get is almost-professional software that only works for almost-professional users. The reality is that Microsoft deserves every penny it gets, because it has designed software that works relatively well for pretty much everyone.
Improbable Joe Thinker Join Date: Feb 2009 Posts: 147

18th July 2009, 09:09 PM	#19
shadron Illuminator Join Date: Sep 2005 Location: Colorado Posts: 3,856	The problem isn't with the fact that software developers and their ~~slave drivers~~ - uhhh, employers shouldn't get paid for the former's work, but rather that the patent system is a blunt instrument for doing that. Rather than having something similar to a copyright law that keeps one's code under the ownership of the writer or his assignee, the patent system is used to freeze the ability for others to use features that could be implemented in many different ways. For example, Microsoft has patents on a number of Windows features which make it impossible for anyone else to duplicate the Windows "look and feel" without Microsoft's license, or the long-running battle over the use of gif image format, which kept that format from becomeing a lot more influential than it really was. No need to take off after Christian for Microsoft's sake, they've amply shown thay can take care of themselves; in fact, if you read carefully, Christian doesn't castigate M$ but rather the American court's upholding patent law for software ideas.

18th July 2009, 09:13 PM	#20
Modified Graduate Poster Join Date: Sep 2006 Location: SW Florida Posts: 1,905	Originally Posted by Improbable Joe "Allows"?!?! Shouldn't software developers get paid, or should they exist as your slave, for your benefit? I'm not a crazy free market capitalist or anything(I tend to mock them) but eliminating the patents on software sounds like an ethical disaster, and the end of all sorts of technological development. Maybe that's what you want: a wasteland where you never pay for anything, and all you get is almost-professional software that only works for almost-professional users. The reality is that Microsoft deserves every penny it gets, because it has designed software that works relatively well for pretty much everyone. Software will be covered by copyright, either way. I can see that maybe a truly novel and as yet unpublished algorithm should be patentable, but that covers approximately zero software patents. Microsoft just applies for patents for everything they do, even if it is copied directly from someone else (http://blogs.zdnet.com/Burnette/?p=245 for example). Many of their patent applications are vague or unintelligible. The "inventors" listed on the patent application probably have never even seen it in most cases. The problem is, any significant software that anyone writes is likely to infringe on several existing software patents, even if the authors believe that every idea and algorithm they are using is unique. For the most part, these patents will never be enforced, and large companies use them mainly for lawsuit defense.

18th July 2009, 09:23 PM	#21
Improbable Joe Thinker Join Date: Feb 2009 Posts: 147	Originally Posted by Modified Software will be covered by copyright, either way. I can see that maybe a truly novel and as yet unpublished algorithm should be patentable, but that covers approximately zero software patents. Microsoft just applies for patents for everything they do, even if it is copied directly from someone else (http://blogs.zdnet.com/Burnette/?p=245 for example). Many of their patent applications are vague or unintelligible. The "inventors" listed on the patent application probably have never even seen it in most cases. The problem is, any significant software that anyone writes is likely to infringe on several existing software patents, even if the authors believe that every idea and algorithm they are using is unique. For the most part, these patents will never be enforced, and large companies use them mainly for lawsuit defense. It sounds like sour grapes to me. There's nothing that is so wrong with Microsoft that the answer is an end to copyright on software, as has been both explicitly and implicitly suggested in this thread. Laws are never perfect, but the idea that the law should be abandoned because people are whining about Microsoft? Not so much... and I'm not even a huge fan of the company. I should correct an earlier statement: Microsoft deserves MOST of the money they get... they can be a giant PITA, but do a good job overall. They can be proud of the job they've done, compared to some companies, even though they have huge room for improvement. My position is that when dealing with these sort of issues, the system isn't the problem, as much as specific vendors in very narrow cases. you need to consider going about it with a laser scalpel, instead of using a chainsaw to cut off entire legal principals.
Improbable Joe Thinker Join Date: Feb 2009 Posts: 147

18th July 2009, 10:42 PM	#22
Modified Graduate Poster Join Date: Sep 2006 Location: SW Florida Posts: 1,905	Originally Posted by Improbable Joe It sounds like sour grapes to me. I'm not sure what you mean. I haven't applied for any software patents, and even if I had I would think they were crap. Quote: There's nothing that is so wrong with Microsoft that the answer is an end to copyright on software, as has been both explicitly and implicitly suggested in this thread. I don't want software patents scrapped to punish Microsoft, I want them scrapped because they are mostly junk. Microsoft would probably be better off without them, since they use them almost exclusively for defense (not to prevent others from using their ideas, but to prevent themselves from being sued for using others' ideas). Quote: My position is that when dealing with these sort of issues, the system isn't the problem, as much as specific vendors in very narrow cases. Then you don't know anything about it. As I said, any significant software that is created is likely to violate multiple patents. Determining which patents may be violated would cost many times the software development cost. Creating many common classes of software without violating patents is impossible. Imagine there were thousands of laws prohibiting common activities that you were unaware of and were rarely or never enforced, but for which you could potentially be punished and which in many cases you could not avoid violating.

19th July 2009, 05:53 AM	#23
laca Muse Join Date: May 2009 Posts: 739	Originally Posted by Improbable Joe "Allows"?!?! Shouldn't software developers get paid, or should they exist as your slave, for your benefit? I'm not a crazy free market capitalist or anything(I tend to mock them) but eliminating the patents on software sounds like an ethical disaster, and the end of all sorts of technological development. You are confusing patents with copyright. What if Van Gogh patented a color so noone else could use it? Mind you, not the batch he made, but the color itself. Now that is what ends "all sorts of technological development". In that kind of world there can never be a painter who sees one of Van Gogh's paintings, loves that color and tries to reproduce it. Originally Posted by Improbable Joe Maybe that's what you want: a wasteland where you never pay for anything, and all you get is almost-professional software that only works for almost-professional users. Strawman. Nobody said that. Originally Posted by Improbable Joe The reality is that Microsoft deserves every penny it gets, because it has designed software that works relatively well for pretty much everyone. Does Microsoft also deserve to be the target of patent trolls?
laca Muse Join Date: May 2009 Posts: 739

19th July 2009, 05:59 AM	#24
laca Muse Join Date: May 2009 Posts: 739	Originally Posted by Modified Software will be covered by copyright, either way. I can see that maybe a truly novel and as yet unpublished algorithm should be patentable, but that covers approximately zero software patents. Hmmm... How about patenting mathematical theorems? To me, that is not so much different from patenting algorithms. Originally Posted by Modified Microsoft just applies for patents for everything they do, even if it is copied directly from someone else (http://blogs.zdnet.com/Burnette/?p=245 for example). Many of their patent applications are vague or unintelligible. The "inventors" listed on the patent application probably have never even seen it in most cases. Maybe. But not only Microsoft does it. Almost everyone does it. There are companies that specialize in patents. That is they register patents in the hope to extort some money from them in the future. Originally Posted by Modified The problem is, any significant software that anyone writes is likely to infringe on several existing software patents, even if the authors believe that every idea and algorithm they are using is unique. For the most part, these patents will never be enforced, and large companies use them mainly for lawsuit defense. Agreed 100%.
laca Muse Join Date: May 2009 Posts: 739

19th July 2009, 06:01 AM	#25
laca Muse Join Date: May 2009 Posts: 739	Originally Posted by Improbable Joe It sounds like sour grapes to me. There's nothing that is so wrong with Microsoft that the answer is an end to copyright on software, as has been both explicitly and implicitly suggested in this thread. Would you mind pointing out who suggested to end copyright on software in this thread? I can't find it and I would have some stuff to say about that as well...
laca Muse Join Date: May 2009 Posts: 739

19th July 2009, 07:49 AM	#26
Christian Klippel Muse Join Date: Feb 2008 Location: Ruhr Area in Germany Posts: 653	Originally Posted by Improbable Joe "Allows"?!?! Shouldn't software developers get paid, or should they exist as your slave, for your benefit? I'm not a crazy free market capitalist or anything(I tend to mock them) but eliminating the patents on software sounds like an ethical disaster, and the end of all sorts of technological development. Maybe that's what you want: a wasteland where you never pay for anything, and all you get is almost-professional software that only works for almost-professional users. The reality is that Microsoft deserves every penny it gets, because it has designed software that works relatively well for pretty much everyone. I think that you have no grasp about the issues at hand. Software is covered by copyright protection already, and that in fact is the proper way to protect it. Maybe you are not aware of the difference between a patent and between a copyright. And before you go of and try to tell me that i want to smear or punish Microsoft, or that i want every programmer to not be paid for her/his work, go back and read what i wrote. The problem with software patents is that software is not a physical thing. It is not an object that has a dedicated function and needs a special way to be produced. Software is much more like literature. You use common words, and lots of them, to create a program. Pretty much the same way that you use a lot of words to write a book. There are many, many things in software that simply can not be done in a different way. Well, maybe they could, but that would result in an really awkward pile of crappy code. Did you know that the progress bar, that virtually every user interface uses, is actually patented? Look here. Take a look here for more absurd software patents. Again, software is already protected by copyright. And that is a good thing. The nastiness starts when software is patented, because that is simply a stupid thing to do. Nowdays it is virtually impossible to write a program that goes beyond the "Hello world!" example without infringing some stupid patent. This blocks innovation, hinders competition and is a serious threat to the economy. Greetings, Chris
	__________________ Humber-physics 101: The treadmill has no ground equivalent. This means that the belt is not the road, but the Earth. ... That means the belt is also a privileged and unique perspective. If not then the treadmill collapses to the real world equivalent of a real treadmill, with different objects at different velocities in the same frame. Either way, no motion.

19th July 2009, 08:02 AM	#27
Christian Klippel Muse Join Date: Feb 2008 Location: Ruhr Area in Germany Posts: 653	Originally Posted by Improbable Joe It sounds like sour grapes to me. There's nothing that is so wrong with Microsoft that the answer is an end to copyright on software, as has been both explicitly and implicitly suggested in this thread. No, that is not what has been suggested. Read my other post. And while you are at it, read the whole thread again. Originally Posted by Improbable Joe Laws are never perfect, but the idea that the law should be abandoned because people are whining about Microsoft? Not so much... and I'm not even a huge fan of the company. I should correct an earlier statement: Microsoft deserves MOST of the money they get... they can be a giant PITA, but do a good job overall. They can be proud of the job they've done, compared to some companies, even though they have huge room for improvement. As i have said very clearly, this is not about Microsoft alone. It is about the general state of things, and i picked Microsoft simply as a vehicle to demonstrate my point. I explicitly said that this is not a problem of Microsoft in general. Originally Posted by Improbable Joe My position is that when dealing with these sort of issues, the system isn't the problem, as much as specific vendors in very narrow cases. you need to consider going about it with a laser scalpel, instead of using a chainsaw to cut off entire legal principals. You are wrong. When a system allows for such things like software patents, that really threaten the market, then the problem is indeed the system, and not the companies. Companies only do what the laws allow them to do. If the laws allow them to issue software patents, then they will do. And of course they will then go on and use this patents to threaten or even extinguish competition. It is not the companies fault, the fault is with the system. In any case, that hasn't much to do with my OP anyways. While patents on certain file formats play a role in that (see the GIF patent, for example), the basic problem is that with many commercial software packages you have no real control over your data. You are forced to stick to the programs without having a decent way out. If everything goes wrong you are left with a bunch of binary files that you can not read anymore. If such a thing happens to a companies files, i guess that would be a disastrous event. Sure, one can sit down and try to export the files from an old version of a given program, and then import them into a different program. But that task is extremely time consuming and error prone. You surely will loose the formatting, font's, etc. So you would have to re-work them all. Exporting and re-importing databases is an even more complex task to do. IMHO, the only way to avoid such situations is to use open standards that are implemented as open source, because only then you have full access to the data, because of the open-source implementation you have some kind of a reference: namely the way the files were created/written and are read back in the first place. Greetings, Chris
	__________________ Humber-physics 101: The treadmill has no ground equivalent. This means that the belt is not the road, but the Earth. ... That means the belt is also a privileged and unique perspective. If not then the treadmill collapses to the real world equivalent of a real treadmill, with different objects at different velocities in the same frame. Either way, no motion. Last edited by Christian Klippel; 19th July 2009 at 08:03 AM.

19th July 2009, 08:13 AM	#28
Christian Klippel Muse Join Date: Feb 2008 Location: Ruhr Area in Germany Posts: 653	BTW, the software-patents problem is not genuine to the USA anymore. Here in Europe we have the very same problems by now, albeit maybe not that extreme. To make the problem a bit more clearer to people that are not into classical software development, take a classical webshop. Take a look at this page to see what stupid patents affect a simple webshop. Then go here to read about how such patents can become weapons. Now, if you have taken a look at the webshop example, take a moment to think about it. How could the used examples be implemented differently? Is there any way to implement them in a different way at all? These are patents on abstract and very trivial things. Every programmer or web-designer with more than one brain cell will come to the very same solutions for the problems at hand. Greetings, Chris
	__________________ Humber-physics 101: The treadmill has no ground equivalent. This means that the belt is not the road, but the Earth. ... That means the belt is also a privileged and unique perspective. If not then the treadmill collapses to the real world equivalent of a real treadmill, with different objects at different velocities in the same frame. Either way, no motion.

19th July 2009, 08:57 AM	#29
Wudang BOFH Join Date: Jun 2003 Location: Sheffield Posts: 4,412	To be fair to Microsoft, when they splt with IBM over OS/2 and said to IBM "Windows is our ball and you can't play, ner ner". IBM replied with its patent inventory and said "well for starters you can stop using cursors becuase we own that patent". As well as this warehouse full. In IBM Hursley we were encouraged to submit out ideas for patent filing but it was quite a process of technical evaluation etc. I had to pull mine when I found someone beat me to it.
	__________________ Aphorism: Subjects most likely to be declared inappropriate for humor are the ones most in need of it.

19th July 2009, 10:47 AM	#30
Modified Graduate Poster Join Date: Sep 2006 Location: SW Florida Posts: 1,905	Originally Posted by laca Maybe. But not only Microsoft does it. Almost everyone does it. There are companies that specialize in patents. That is they register patents in the hope to extort some money from them in the future. Apart from the patent trolls, most small software companies do not do this. Microsoft does not seem to worry about prior art even when they are well aware of it. Other large companies may be as guilty, but I not seen such obvious cases that got any publicity.

19th July 2009, 10:57 AM	#31
Modified Graduate Poster Join Date: Sep 2006 Location: SW Florida Posts: 1,905	Originally Posted by laca Hmmm... How about patenting mathematical theorems? To me, that is not so much different from patenting algorithms. As I indicated, I have difficulty with the idea. For a truly novel algorithm, something that took significant time and insight to develop but which is difficult to hide from reverse-engineers of the code, it would be nice to have some intellectual property protection for some length of time. Determining that an algorithm is truly novel, significant, and non-obvious to the degree required would be very difficult.

19th July 2009, 11:05 AM	#32
Modified Graduate Poster Join Date: Sep 2006 Location: SW Florida Posts: 1,905	Originally Posted by Christian Klippel Did you know that the progress bar, that virtually every user interface uses, is actually patented? Look here. At least that one is well written.

19th July 2009, 01:27 PM	#33
laca Muse Join Date: May 2009 Posts: 739	Originally Posted by Modified As I indicated, I have difficulty with the idea. For a truly novel algorithm, something that took significant time and insight to develop but which is difficult to hide from reverse-engineers of the code, it would be nice to have some intellectual property protection for some length of time. Again, how is this different from a new mathematical theorem? Originally Posted by Modified Determining that an algorithm is truly novel, significant, and non-obvious to the degree required would be very difficult. A truly novel algorithm that took significant time and insight to develop might be recognized by requiring about the same time and insight to reverse engineer as to develop.
laca Muse Join Date: May 2009 Posts: 739

19th July 2009, 02:54 PM	#34
Modified Graduate Poster Join Date: Sep 2006 Location: SW Florida Posts: 1,905	Originally Posted by laca Again, how is this different from a new mathematical theorem? Because they are two different things? I think you mean how is that different from an algorithm without an intended use in software. If an algorithm is used for some other purpose, then it could possibly be maintained as a trade secret, otherwise, if it is useful in some way but must be exposed in the process, then it should be treated no differently from a software algorithm. Quote: A truly novel algorithm that took significant time and insight to develop might be recognized by requiring about the same time and insight to reverse engineer as to develop. I was talking about reverse engineering the code to discover the algorithm.

19th July 2009, 03:58 PM	#35
laca Muse Join Date: May 2009 Posts: 739	Originally Posted by Modified Because they are two different things? Both are thought processes expressed in some way. And thought should not be patentable. IMHO. Originally Posted by Modified I think you mean how is that different from an algorithm without an intended use in software. If an algorithm is used for some other purpose, then it could possibly be maintained as a trade secret, otherwise, if it is useful in some way but must be exposed in the process, then it should be treated no differently from a software algorithm. Are you suggesting that mathematical theorems have no use? What I'm trying to say is that an algorithm and a theorem are both tools. Neither is an end product. Well, most of the time. Originally Posted by Modified I was talking about reverse engineering the code to discover the algorithm. Well so was I. Been there, done that. It is hard enough to reverse engineer a stupid copy protection scheme, not to mention a complex algorithm.
laca Muse Join Date: May 2009 Posts: 739