Stallman And Bero Interviewed

Juraj Bednar writes: "I have done two interviews: one with Bero from RedHat and one with Richard Stallman, the GNU and FSF founder. I usually write in my native language, but since these interviews were done in English, I asked myself why not to share them" Readers may want to also visit Bero's shared-source.com, and bookmark it as a FUD antidote.

Comments

Re:Free Software, Intellectual Property & Freedom

[osgeek]

The land and its wheath should be divided as an inheritance to be passed to our children and their children. What we do with our piece of the pie is up to us. Demand freedom! Always!

So, how is this division of land being decided, kind of a divide the area of the earth by the number of people on it and we each get a slice of that size?

• Do I get to keep my slice for as long as I live?
• Can I give my kids only my slice, or can I give them pieces of other peoples' slices as well?
• Will there be some sort of redivision of the earth at some point?
• If my next-door neighbor breeds like a fucking rabbit and has like 26 kids on his slice, do they all have to make do with less?
• Can I trade my slice to someone else for some other type of goods or service?
• If I can do whatever I want with my slice and I give it away, where do I sleep? Am I fucked?
• I'm really into PCBs, can I pollute the hell out of my slice?
• If I die and I don't have kids, who gets my slice?

Re:don't forget the other interview!

[bero-rh]

First of all, this is not some generic interview section - it's just a pointer to some questions about "shared source" I answered a while ago.

But since I'm reading this, guess there's no reason not to reply. ;)

1) Ever since the Qt license problems have been resolved, RH hasn't had problems with KDE. Actually, most people in this office use KDE.
There have been a couple of internal flamewars of course, but nobody really takes them seriously.

2) Sure - some of the most serious gripes I've had with Red Hat Linux when I started BeroLinux have been fixed for quite a while - for example, the lack of a possibility to add a non-root user during installation (added in 6.1), KDE integration (initially added in 6.0, updated to a sane version in 7.1), or wasting space by not compressing man/info pages (fixed in 6.1 or 6.2, don't remember), or the lack of optimizations (all 7.x releases are compiled with -march=i386 -mcpu=i686). There are still some things I'd do differently, but overall, I'm quite satisfied with the current version (the current beta in particular).

3) Yes, to an extent. It annoys me even more that RH never bothered to make an official statement regardning the compiler.
I think the whole thing wouldn't be the way it is if someone in power had taken the time to communicate it correctly, preferrably before the 7.0 release.

4) That strongly depends on what you want to do - I personally want to eliminate the need for non-free OSes, which means usability (and thereby KDE) needs the most attention at the moment. But then, things like scaling down to embedded devices and up to high-end servers are not exactly useless either... I think going ahead in all directions the way it's happening now is a good thing.

5) We have a more generic approach to prelinking (needs a patched ld.so and binutils though). This is part of the current beta of RHL.

Re:Free Software, Intellectual Property & Freedom

[Alpha State]

There is value in the capitalist system, I'm sure most people would agree. To me the value is this: in the capitalist system if I contribute to society by producing something of value, I make money. If I am intelligent and work hard I can make a very good living, even become "rich". Those who are too lazy to contribute do not make money. Thus there is a very high incentive to do something worthwhile. Of course, this is in theory and there are plenty of holes in the system, but it does work.

If there is no intellectual property, the capitalist system will not work for it. Thus there will be no incentive to work on IP and loafers will get a free ride. I do not believe this really appplies to free software because it is produced by cooperation between people who need the software, I write a program because I want to use it and share it because I wish to, and it may make the program better.

There are other ways to provide incentive for IP, such as the above, or commissioned work, or street performer protocols, etc. But they won't work for every kind of IP, and there will be big problems in integrating with the capitalist system.

I agree that IP laws are becoming more draconian, but before the relatively recent WIPO treaties and associated laws there was a fairly good balance between the needs of IP producers and consumers. What will happen when this balance is disturbed? I predict that IP consumers (ie. the general public) will become more and more willing to break the restrictively laws. It could end messily unless the laws are changed, just like most regimes who have sought to enslave their citizens.

Hack this box.

[Faux_Pseudo]

If you read the whole shared-source.com page there is a section on "if open source is so unsafe then hack this computer". It lists all of the software and version numbers the box is running.

It would be nice if a MS website where able to be that bold.

Who is to write software, then?

[bwalling]

I'm serious about this - I'm not trying to be a troll.

If software is to be free, then who can we expect to write it. Obviously, I have a need for a paycheck. Since I have this need, I have an employer. In order for my employer to pay me, I have to contribute to their revenue.

Is it reasonable for companies to only make money from services, and to offer the software for free? Are there companies who are successfully doing this? (Yes I saw the RedHat Quarterly report, but that was a little number fudging - they still lost money). Do we just have to wait out a certain transition period before the idea of Free Software pervasively existing is realistic?

Re:Who is to write software, then?

[Chris Johnson]

Easy: people.

What gives you the notion that writing software must be constrained to programmers?

At http://www.airwindows.com/dithering/index.html you will find an elaborate program for high-end digital audio mastering from >16bit word lengths. It includes a number of very killer vertical-market type features like multiband sidechain compression. It does NOT have remotely professional file-reading and writing, because those are more 'real programmer' things, and I'm not a 'real programmer'. However, no 'real programmer' has shown any interest in writing such an app, and the market is so tiny that the few people building stuff for it tend to charge in the kilobucks- and the app I did is GPLed and just to have it costs nothing.

So it is not a question of 'so if you wanted said mastering software, how would get it if nobody will write it without money?'. Surprise! Nobody wrote it anyway. The 'market' did not lead to any such software existing, even though I needed it desperately.

And it is not a question of 'yeah, right, like a programmer is going to do hard work like that for free': clue jet coming in on runway six, a programmer didn't do that. I did. It's not done in the way you'd want to sell as shrinkwrapped greedware, but then the market's too small anyhow. The point is, this program _exists_ and grows and evolves based on just one person's ability to mostly sort of program. It's GPLed making it that much easier for the _next_ person who has a personal task to accomplish, to get a head start. And that's how it goes...

I really have little patience for programmers. Programmers are like the people who put the spyware boobytrap 'dial up and invalidate the registration number if the person's reinstalled the program too many times' code into an mp3 player app that I _bought_ and ended up demanding my money back on. There's a lot that you don't really need a programmer for- you need one for good games, for serious server apps, for the _computery_ stuff, but there's a million other things that can be done more crudely by just regular people with a bit of determination.

(I'm not _really_ against programmers- not like that- but I grow very sick and tired of the 'software can't be free, how will you survive without paying US?!?' refrain. Maybe you're not as indispensable as you think.)

They did give a challenge - two years ago.

[DeeKayWon]

Right here.

Re:RMS phoned it in

[stevens]

But it looks like Stallman just copied and pasted some boilerplate.

Well, he wasn't asked a question he hasn't answered five hundred times. I'll bet I could find decent answers to every question in that interview on the fsf website.

If you want novel and thoughtful responses, ask novel and thoughtful questions.

Bero not quite accurate about GPL and derived work

[joneshenry]

I'm sorry but I don't believe that Bero is being completely accurate when he claims that the "GPL makes no claims to data generated, processed, or stored by something covered by it." According to the text of the GPL "The act of running the Program is not restricted, and the output from the Program is covered only if its contents constitute a work based on the Program (independent of having been made by running the Program). Whether that is true depends on what the Program does." IANAL, but in my opinion, for a company to be completely safe about using the output of GPLed software, they must examine every line of the source code. The reason is that it is possible that the program will inject portions of itself into the output. An example is Bison whose license was modified. To quote documentation from an older version of Bison 1.20 "Bison grammars can be used only in programs that are free software. This is in contrast to what happens with the GNU C compiler and the other GNU programming tools. The reason Bison is special is that the output of the Bison utility--the Bison parser file--contains a verbatim copy of a sizable piece of Bison, which is the code for the yyparse function. As a result, the Bison parser file is covered by the same copying conditions that cover Bison itself and the rest of the GNU system: any program containing it has to be distributed under the standard GNU copying conditions." The license was later changed in version 1.24 and beyond: "As of Bison version 1.24, we have changed the distribution terms for yyparse to permit using Bison's output in non-free programs. Formerly, Bison parsers could be used only in programs that were free software."

This is a common flaw in thinking.

[dmaxwell]

There is a certain prevalent mentality that assumes the only significant motivation for doing anything is the desire to make money. There are a great many craftsmen (and women, I know) who would not say money is their prime reason to be doing what they are doing. Most of the best art falls in this category. The stuff made primarily for BIG SALES tends to suck. Britney Spears anybody?

Some people write this stuff because it is fun to run their own code. Others do it because the software in ancillary to their true goals. The Apache web server came about this way. Apache wasn't developed to make Webserver Inc a pile of money. Some webadmins needed a httpd daemon that was reliable and featureful. The original Linux kernel that Linus made available to his fellow hackers wasn't going to make anybody mounds of cash: it would barely boot a 386. The additions from volunteers was what made it valuable.

I'll agree that anybody who wants to make money trying to sell something that is free is on a fools errand. However there is nothing wrong in taking something free and using it as part of something larger that is sold. The school district that I work for uses a product called the Firebox. It is not marketed strictly as a Linux box. It is sold as an easily configured firewall and proxy server. The middle school tech guy loves that thing. Oh yeah, and they pay the guy who works on iptables. IBM and SUN are hardware companies and are all for anything that helps them sell hardware. Incidentally, the bulk of RedHat's profit doesn't come from selling the boxed distro. They also sell customization and consulting services.

Open source only fails to make sense to those who sell boxed software. It is a moneymaking or moneysaving opportunity for others with different models. Think about independent music for a monent. With the RIAA gone there wouldn't be many pop music multimillionaires. There would be and ARE a lot of people who earn honest livings writing and performing. The same is true of open source. No one will be a multibillionaire selling it but it will enable many others to earn decent livings.

don't forget the other interview!

[reaper20]

I'm glad we finally get an interview with bero, one of the most underrated hackers out there imo, i've always been fascinated with the Bero Linux distribution from a while back and everything he's done with RH so far. I would have asked some other questions, so since we all know he reads slashdot I have a few questions:

1) You apparantly host dot.kde.org and post regulary, though RH 'sponsors' GNOME. Anyone at redhat have any comments towards you? Hate mail? Unexpected nerf ambushes? Do they sign you up for GNOME mailing lists? Do they make fun of the 'KDE guy in the corner'? (BTW, this is what makes open source so cool, the freedom to choose what you want).

2) Any of the ideas from bero linux make their way into the main RH distro? I know Mandrake did, but since RH is mostly conservative, I'd like to hear your opinion.

3) Does it piss you off that every complaint about the gcc in RH is answered on your website and you have to post the URL for the last 2 red hat releases including the betas? (that must suck).

4) You get paid to work on Linux, that rocks! What do you think needs the most attention?

5) Any chance that prelinking stuff will make KDE2.2? How about any of the other RH packages?

Thanks, and thanks for the kde daily builds ... they rock.

Re:danced around the communism question

[Chagrin]

There is a huge gap between communism and socialism. The GNU manifesto follows much more closely to socialism.

Socialism: Any of various theories or systems of social organization in which the means of producing and distributing goods is owned collectively or by a centralized government that often plans and controls the economy.

Communism: A system of government in which the state plans and controls the economy and a single, often authoritarian party holds power, claiming to make progress toward a higher social order in which all goods are equally shared by the people.

I don't see where we have any authoritarian parties holding power, so please don't compare the GNU movement to the Soviet system of government.