Richard Stallman on The Setup | Hacker News Reader

Richard Stallman on The Setup

richard.stallman.usesthis.com

70 pointsmqt103 comments

I don't believe he uses only his netbook. I think he also uses the public CSAIL computers at Stata Center in MIT. They have a KDE Linux on them.

I have used them from time to time, and as before logging in, at the username prompt usually you can see the last user that was on the computer. Many times it was "rms".

Typical Stallman... taking everything to the extreme and suffering the consequences. It's the burden of the idealist.

I admire his willingness to stand to what he believes.

Also, extremisms aside, he's contributed a lot to the community.

Typical Stallman ... having principles and living by them.

Beliefs are the bastion of the uninformed, and those so set in their ways they can't take a moment to think things through.

What you call beliefs, I call principles. It's nice to know he doesn't compromise them.

Why is that a nice trait to have?

The world isn't perfect. It's certainly not black or white either. You have to compromise.

Often it's the people that refuse to compromise that make the world a better place. Do you think Martin Luther King should have compromised on his vision of racial equality?

Whilst the civil rights movement is obviously far more important than the free software movement, the idea that ideals should not be compromised is at the core of both. Unlike, say, the BSD license, the GPL is a license that allows for no compromise. If you want to distribute modifications, you must also distribute the source. It's this ideology that made the more pragmatic open-source movement possible in the first place.

How is Richard Stallman refusing to use a webbrowser making the world a better place? :/ It's just taking ideals to extremes for the sake of it.

As far as I'm aware, Stallman isn't refusing to use a web browser for ideological reasons. It's because: "Most of the time I do not have an Internet connection."

Besides, I seem to recall he uses Emacs/W3 for viewing web pages (offline, downloaded via wget).

>> "To look at page I send mail to a demon which runs wget and mails the page back to me."

You have to admit that sounds slightly quirky? I assumed it was an attempt to maintain that his personal PC is free from any closed source software, since it's not even rendering websites, it's just receiving email.

Sorry, but to me it sounds like either someone who is so into ideals that they're cutting off their nose to spite their face, or someone who has simply failed to keep up with changes in technology.

No, I'm pretty sure that is an attempt to remain mostly-distraction free and not waste time on the internet. There are (several) entirely open source/free browsers he would use if that was the issue.

The internet does have some useful productive things on it as well as time wasting devices :/

For a start it's pretty useful for documentation.

That's what wget -r is for.

Pretty sure Stallman isn't going to show up in this thread ;)

According to the interview he only has an internet connection two or three times a day at most. In that sort of scenario his set up kind of makes sense. If I only had short bursts of internet connections during the day I'd probably also set up a system where I could queue up URLs to be automatically downloaded for offline reading next time I had a connection.

You're missing the point. why does he have such limited connectivity?

Whatever the reason I doubt it has anything to do with closed software or his free software philosophy. There are plenty of completely Free ways to get online. Perhaps he lives in a place where you can't get broadband, and only bothers to dial up a couple of times a day. Or perhaps he simply doesn't want a permanent Internet connection because he doesn't feel a need for it.

Actually there have been at least a couple of threads about working without internet on a some of the RV/nomad threads here, and based on those I'm considering trying pulling the plug on my internet connection for 4 hours at a time when working to see what it will do be my ability to focus on the task at hand and not get distracted by hacker news. Perhaps Stallman has a similar approach.

When I have a tight deadline, blocking Google Reader, HN and Slashdot increases my productivity by a couple orders of magnitude.

> Most of the time I do not have an Internet connection.

He should get it then and join us all in the 21st century.

Because there are (a lot of) people on the close-everything side of the debate it's essential to have at least a few people arguing that everything is opened so that the eventual compromise is somewhere in the middle.

I refuse to call it GNU/Linux. By pure lines of source it should be called X.org/Linux.

Why X.org? Even if you ignore the fact that it is an XFree86 fork, there are plenty of Linux systems with no X at all. In fact, that would be most of them, since there are certainly more Linux servers than desktops out there. Now add the embedded systems, and desktop installs are a drop in the bucket. Also, a number of GNU tools predate the Linux kernel.

Apparently HN is refusing to let me reply to you. So I'm doing it here.

The point of saying X.org/Linux is that I could form some ridiculous argument that would require a completely different naming convention. Why don't we call all cars Benz/Cars since Karl Benz holds much of the work that makes cars go? Or all helicopters Da Vinci/Helicopters? Because it's silliness.

Calling something GNU/Linux is silly nonsense RMS designed so he could get his beard stroked. If you notice he's never suggested the following:

-Linux (powered by GNU)

-Linux-GNU

-Linux/GNU

-Linux - With GNU utilities

-Linux - powered by GNU and Xfree86/X.org and the Mozilla foundation.

-Linux/GNU/Xfree86|X.org/Mozilla

- etc.

He's suggested only GNU/Linux. Notice GNU is in front (to remind us all how wonderful GNU is) and the '/' indicates it's interchangeable. Like "do you want ham and/or cheese on your sandwich?". Only GNU != Linux, the two are not interchangeable. The GNU Hurd is a failure. But I bet I could cobble together a GNU free Linux distro for the most part today, even rewriting most of the utilities in a couple weeks, and probably completely in a couple of years once LLVM gets up and going.

He wanted to co-opt the relative success that Linux became by formulating an argument such that the only resolution in his mind was to call it GNU/Linux. And instead of building his own system, he built a bunch of the crap for a completed system, then spent the rest of his years dwelling in Emacs and emailing himself wget harvests of websites because for "personal reasons".

Really, outside of maybe the compiler toolchain (which is looking to have a short lifespan in the face of LLVM and associated front ends like Clang, are any of the utilities that hard to replicate? (quick answer no, I had to build most of the GNU toolkit from scratch for a single OS course assignment in my undergrad, including a shell, everybody had to do that, it's pretty easy). I'm sorry, but a few dozen utilities like 'ls' and 'cat' don't allow somebody to co-opt somebody else's hard work.

Actually it makes perfect sense. Most Linux systems are a collection of GNU tools on top of the Linux kernel. This "bunch of crap" forms the basis of a lot of internet infrastructure you take for granted. Also, saying the GNU compilers have a short lifespan is, well... Perhaps they do, in geological terms. Seriously, GCC existed since like ever.

My personal setup could be called Gnome/Xorg/GNU/Linux, but if people complain about GNU/Linux I guess my letter soup won't fly. Debian is packaging something that is called GNU/kFreeBSD, a GNU userland on top of a FreeBSD kernel. Nexenta could be called GNU/OpenSolaris (in fact, there are lots of GNU tools in OpenSolaris right now).

Operating systems, from kernel to shell, are usually conceptually very simple, with tons of hard work on top of simple concepts. This simplicity, specially in the case of Unix, allowed it to be continuously reinvented for 40 years in a way it's still a modern OS. I really wish we had HAL/OS (maybe Microsoft can expand the Milo and Kate demo into one) and full-blown human-level AI agents, but the world didn't take any turn into that direction. Unix, in its many incarnations, is still the OS to imitate. And the GNU tools, be it on top of a BSD/Mach kernel (as in OSX) or the Linux kernel is more or less what comes to mind when someone says "Unix".

That is quite an impressive accomplishment.

> Seriously, GCC existed since like ever

Absolutely, I remember using the GCC tools way back when. Not disputing that. But there are certainly older (and arguably better) compilers. And there will be newer ones (like the upcoming LLVM setup). GCC does not hold a monopoly on converting and optimizing high-level language constructs into localized machine code.

> My personal setup could be called Gnome/Xorg/GNU/Linux

Which is fine. Either call it "Linux" or call it a string of alphabet soup if you like.

My problem with RMS is his insistence on only GNU/Linux. Which is absolutely disrespectful to the hard work Linus and others put into the Kernel, which RMS and the GNU foundation was unable to accomplish, and to the folks who've written numerous other things that are the bread and butter of what many people think Linux boxen are for - like the Apache Foundation or XFree86/X.org or the KDE folks or Sun or the various thousands of volunteers that have toiled away writing drivers or firmware or whatever...It's a great exemplar of near infinite hubris to insist it be "GNU" (first) / Linux.

From wikipedia (and I've seen similar quote from him over the years) "Stallman argues that not using "GNU" in the name of the operating system unfairly disparages the value of the GNU project..."

It's unfortunate that he doesn't understand co-opting the work of others under your own umbrella is what's disparaging the GNU project's name. I don't think anybody who is marginally familiar with the topic thinks that the GNU foundation isn't an important thing. But do I have to stick GNU on the front of anything I build out of GNU stuff? What if I decide to write a cookbook about French Cooking? Do I have to call it the "Houghton Mifflin/French Tarts for Tarty People" since Houghton Mifflin published Julia Child's books which I used to learn the French culinary arts?

That's silliness.

Linus happily acknowledges the role the Foundation has played, with very little prompting. The credit is already being granted.

http://www.youtube.com/watch?v=Z8ugRM3-twc

But he also says

"Well, I think it's justified, but it's justified if you actually make a GNU distribution of Linux ... the same way that I think that "Red Hat Linux" is fine, or "SuSE Linux" or "Debian Linux", because if you actually make your own distribution of Linux, you get to name the thing, but calling Linux in general "GNU Linux" I think is just ridiculous."

And this same feeling is voiced by others like Jim Gettys

"There are lots of people on this bus; I don't hear a clamor of support that GNU is more essential than many of the other components; can't take a wheel away, and end up with a functional vehicle, or an engine, or the seats. I recommend you be happy we have a bus."

and other unhappy voices in the community including Eric S. Raymond,

http://en.wikipedia.org/wiki/GNU/Linux_naming_controversy

It all boils down to sour grapes which the Linux Journal correctly identified in this single summary statement

"Linus got the glory for what [Stallman] wanted to do."

And what you call principles, I call dogmata.

Why not compromise and call them principles, just this once? Principles, dogma-- what's the difference, really?

Because if you compromise they're not principals...ok, bad joke. Mod me down.

I'll mod you up for the joke, but down because it was the same as the one I made.

Damn, I missed yours it was so subtle.

Why?

Because that's what they are.

From Wikipedia: "Dogma is the established belief or doctrine held by a religion, ideology or any kind of organization: it is authoritative and not to be disputed, doubted or diverged from."

Some examples:

* 4 rights constitute Free Software, not 3 or 2 or 1 but 4.

* Free Software is equivalent to Civil Freedom and Human Rights.

* Distributing software without any of these 4 rights makes the act unethical, immoral and evil.

* All software should be Free Software.

Doubt any of these and you are no true believer of the Free Software Movement.

The Free Software Foundation defines the traits of Free Software as having 4 freedoms (not rights) and explains them in simple language at http://www.gnu.org/philosophy/free-sw.html.

The freedoms are enforced by the GPL, just like any other EULA. So, trying to question why there are only "4" principles, is missing the point. The GPL itself has several terms and conditions.

So, equating the 4 freedoms to the notion similar to the "religious 10 commandments" and then invoking the argument of "dogma" seems disingenuous.

First time I've come across the setup. Here are some interviews with a few other HN regulars:

http://paul.graham.usesthis.com/ http://aaron.swartz.usesthis.com/

_why is listed on their mainpage too but his interview seems to have been removed

_why's link works fine: http://_why.usesthis.com/

Drawn and quirky (or precious, depending on your taste or mood) as normal for _why.

Only if you use a noncompliant DNS server. Only letters and digits are permitted to start a hostname.

http://www.apps.ietf.org/rfc/rfc1123.html#sec-2.1 http://www.apps.ietf.org/rfc/rfc952.html http://www.apps.ietf.org/rfc/rfc921.html

Hmmm. I think this depends on your DNS provider. This link is broken for me: I think my ISP does not like hostnames starting with underscores.

Interesting. If it helps or anyone else is curious, I'm using OpenDNS.

This is really strange. None of the browsers on my machine can access this URL. But

  dig _why.usesthis.com

returns the correct info.

According to Wikipedia, the use of underscores in hostnames is illegal, but many implementations ignore this. Anyway, the interview is mirrored here: http://viewsourcecode.org/why/#7

http://www.apps.ietf.org/rfc/rfc1123.html#sec-6.1.3.5

Looks like underscores are valid in /domain names/ but not in /host names/.

You'll see underscores used many times for SPF records and in Microsoft's Active Directory.

Also, when you use dig I'm pretty sure it manually recurses the name server hierarchy while an application will just connect to your local name server which might have its own policies.

It sounds as if he operates in batch mode with regard to communication.

In 2007 at least, he opted to use a demon to email him pages rather than browse the web:

    For personal reasons, I do not browse the web from my computer.  (I
    also have not net connection much of the time.)  To look at page I
    send mail to a demon which runs wget and mails the page back to me.
    It is very efficient use of my time, but it is slow in real time.

http://lwn.net/Articles/262570/

I hope it's a joke. I actually did this some 12 years ago. We only had email access at my high school, and I used to browse ftp sites by email. With emails being sent only twice a day. I leaned pretty quickly that all the good stuff is in /pub... saved me a full day of browsing.

He really is >this< close to being certifiable.

"I also have not net connection much of the time"

Even that's quirky, when you can hardly buy a phone without Internet capability these days. You have to be deliberately avoiding a connection.

Those phones you speak of, do they run 100% Free Software? If not, rms isn't interested.

Is it really? Don't you get sick of being "always on" sometimes?

Perhaps the averse reactions here have something to do with too many of us living in densely populated/over-populated or even over-stimulated environments; the short-attention-span generation?

Sometimes I do... I give my self breaks, I also be sure to not watch television and read books often. It helps.

Even when you are off the grid, you can still get 24/7 high-speed internet with bi-directional satellite!

Sure, but it doesn't mean I'm without a network connection. There is no situation that I'd want to be uncontactable by everyone. I've an IM client on my phone that a very few people know the account, that will always notify, even if I've chosen to ignore every other kind of message.

RMS is just being ornery.

RMS had actually switched to an OLPC XO laptop but then dropped it:

http://www.bostonreview.net/BR33.6/stallman.php

Does anyone know of a US distributor for the Yeeloong netbook?

+1, that is an interesting article. I read he dropped the OLPC because it allowed running Windows on it. He expects the OLPC to turn millions of children into Windows users, and compares Windows to an addictive drug, and Microsoft to a drugs dealer supplying kids with a cheap first 'fix'. Also, Windows will not run on his Lemote laptop.

I will personally buy a Yeelong netbook and install Windows on it just so RMS has no computer to use at all if that's how he swings.

"This interview is available under the Attribution No Derivatives license."

Am I the only one who finds that ironic?

The reason for "No Derivatives" is that RMS doesn't want his words to be changed. Some text files inside Emacs are (c) RMS, but with a permission to copy and distribute the entire article verbatim.

Not especially, Stalman campaigns for free software and he has reasons why he thinks it's a very good idea. Those reasons don't really extend to interviews as far as I can see.

That's because despite his dogged persistence on the FSF GNU bit, he really doesn't care about Freedom.

You think what he has quite literally dedicated his life to doesn't count simply because he didn't grant you permission to put words in his mouth?

I think RMS is restricting my freedom to put words in his mouth.

I don't understand the god like status given to this guy. He's clearly crazy and from a whole other world than the one real people live in.

Who really cares if their BIOS is open source or not? Why does it matter? Are the designs used to make his laptop case open source as well? I doubt it.

http://www.virtualteacher.com.au/crazyone.html

Seriously though, rms, is one of the principal founders of the free software philosophy, and creator of the GPL, upon which some significant software has been licensed. He's consistent in his vision, which is freedom, and can reliably be looked to for guidance on topics regarding those freedoms.

Can you identify even three people who have had as significant impact on free (as in speech) software?

[edit - I don't mean people like linus, who create great software, but people who develop/evangelize free software philosophies - Theo De Raadt comes to mind - there aren't too many others that I can think of right away]

Free software existed before the GPL. Pretty sure linux would still have happened without it.

People are moving to a web based architecture all the time. The amount of software people actually download and install is reducing, so to some extent the whole idea of OSS for most people (I believe) is becoming less important.

Meanwhile Richard Stallman doesn't even use the web.

I just don't see it as that relevant anymore.

I have a feeling it's only going to get more relevant, because now the software is moving toward the web, it is insanely convenient and easy to lock-in users and restrict freedoms.

There are of course a freaking huge number of problems with the idea of open-sourcing the entirety of many website backends. I think there can be great discussion though on various models. GitHub offering web hosting instead of just repo hosting seems like a possibility.

>> "it is insanely convenient and easy to lock-in users and restrict freedoms."

I agree, but is that really a bad thing? We all have to make money somehow. I'd rather website owners had real incentive to please their users, and being able to make money is a good incentive.

I don't see how open source can really be relevant on the web. If facebook want to do something to drive users away, or ristrict users freedom, that's up to them. If they do it enough, someone else will setup a rival website that pleases users better. How does open sourcing fit into this? I don't think it does. Most website is simple enough that you don't need free source code. Any one of us could clone facebook if we had enough time and determination.

Open sourcing is certainly a nice thing to do, and I'm sure we'd be lost without our own favorite open source tools, but I think it makes less sense on the web.

As I've said before, I don't use X because it's open source, I use it because it's free, and there are people behind it who care.

I think that's all that matters really.

With the (welcome) move to web architecture, free software is not as important as free data, i.e. the freedom to move your data to another provider. We may all be slaves, but we should have the freedom to choose our master.

Like you, I would also rather that website owners had real incentives to please their users. I disagree on the solution: data lock-in _removes_ the website owner's incentive, instead they can just rely on the fact that you are stuck with them, rather than pleased with them.

Good point, and I agree. People should always be free to get at the data they have on a website, and move to an alternative website. That's certainly something worth fighting for.

Is locked in data really a bad thing?

We all have to make money somehow. I'd rather website owners had real incentive to please their users, and being able to make money is a good incentive.

I don't see how free data can really be relevant on the web. If facebook want to do something to drive users away, or ristrict users freedom, that's up to them. If they do it enough, someone else will setup a rival website that pleases users better. How does free data fit into this? I don't think it does. Most website is simple enough that you don't need free data. Any one of us could clone facebook's massive social network if we had enough time and determination. Letting users own their data is certainly a nice thing to do, and I'm sure we'd be lost without our own data, but I think it makes less sense on the web.

As I've said before, I don't use X because its data is free, I use it because it's free, and there are people behind it who care.

I think that's all that matters really.

I don't see how you seem to equate users data, with source code. The two are entirely different things.

Locking in users and their data removes incentive for developers to please their users. If you know that all your user could cancel their service contract and be set up with a competitor within an hour without losing anything, you have a damn fine incentive to keep those users happy. If on the other hand moving to a competitor means losing access to years of data and weeks or month of work just to get back to where they started then you have to suck pretty damn hard for people to leave you.

The (A)GPL is still very relevant. The whole point of free software (in the GNU sense) is that you should have complete access to the sources of a software system.

In the case of a web application, this includes server side software. No matter what some vocal AGPL opponents here say, this opinion is not exclusively held by fundamentalists:

http://blogs.zdnet.com/BTL/?p=15855

This is true,

I think RMS is as close to certifiable as one can get, but the point of having source available for inspection is a very important one in certain contexts. A quick example, voting machines, for quite a while the electronic voting machine sources were completely closed and the companies fought in court from having to reveal the source. When the source was finally reviewed it was found to be a mess and could possibly turn an election one way or the other.

http://www.pcworld.com/article/135461/diebold_voting_machine...

This is one of those areas where Freedom and Free Software intersect in a way where Stallman's philosophy makes a heck of a lot of sense.

On the flip side, not browsing the web for "personal reasons" sounds like somebody who also has tin foil blocking all their windows.

Oh don't be silly, "real people"? Richard Stallman is as real as it gets. He's a decent programmer from what I can tell from emacs.

Note that there is a distinction between open source and free software. I think it's important to not conflate the two, which I see a lot these days (in the real world :)

I predict 500 years from now people will remember him. They won't remember Bill Gates. Without googling try to recall who was the richest man in Rome?

All the best

"I predict 500 years from now people will remember him. They won't remember Bill Gates."

That future is going to spring from this present, where N people use Windows and Word and know who Bill Gates is, but M people use emacs and know who Richard Stallman is?

We see 2 very different futures then.

I see a future where people just access 'the web', via whatever small thin client web browser device they like.

The vast majority of software will be running on servers, and end users won't really care if it's open source as long as it does what they want it to.

Real people use web browsers.

we've been to this future before, when mainframes ruled. We called them dumb terminals. Not as sexy as small thin web clients using comet and ajax and other cleansers but they worked.

kidding aside, I don't buy it. P2P with local datastores will have a place. There's no way my tax returns will reside on google docs.

Your tax returns will reside on the governments web interface :/

In the UK practically nothing tax related is done via paper forms any more, it's all done via their website. Works pretty well.

Crassus?

Beat me to it. Maybe Trajan, depending on how your accounting systems attribute what belonged to the Republic and what to Trajan the man/military commander/later emperor.

http://en.wikipedia.org/wiki/Trajan#Maximum_extent_of_the_Em...

well done. Admittedly I've read parts of Gibbon and still don't recall this.

If you go up to any person on the street today and ask who Bill Gates is, they'll more than likely know who. You average person would have no idea who Richard Stallman.

If your average person doesn't know who he is today, I doubt that'll change to being more noteworthy in 500 years.

The first part is true, but in general the second part is false; in general, there is a big difference between the people who are most famous in their lifetimes and those who are most famous later from that period.

Exactly so. George Brandes was orders of magnitude more famous than Nietzsche in the 1890s. There are countless such examples.

Is there a name for the branch of history that deals specifically with "pop culture" of past societies?

Not everyone knows who Bill Gates is.

http://www.geeks.co.uk/12352-bill-gates-is-a-thief-and-a-don...

GNU is a lot more likely to be around in 500 years than Microsoft or Windows.

Possible. But no one will know/care it's GNU.

I don't see any evidence that he is crazy.

This reminded me of the Don Dodge GOOG/MSFT transition post. Maybe it was the displays of platform loyalties, or their perceived backwardness of the technology habits?

http://dondodge.typepad.com/the_next_big_thing/2009/11/thank...

Stallman is pretty steady and unchanging in his habits towards software and internet access.

I really am not familiar with this man, but comments about his unwillingness to compromise and questioning his sanity make me think of this quote:

The reasonable man adapts himself to the world; the unreasonable one persists in trying to adapt the world to himself. Therefore all progress depends on the unreasonable man. -- George Bernard Shaw

Pretty true indeed - as long as the mad man DOES something. As far as Stallman goes, he coded emacs (along with other people, afaik), gave a name to a movement - and lived ever after on complaining, criticizing other people and complaining some more. If we'd depend on this crazy dude for progress, we'd all be using "fully open-source" inferior hardware produced by kids in China. Oh, and most of us would be jobless too, since NO ONE PAYS A DIME FOR FREE SOFTWARE! Talk about progress...

accidentally touching the mouse-pad and moving the pointer, which would be a nuisance

Sheesh. One would hope he has better reasons than that for using console. It's not that hard to configure X to not switch window on mouse-over.

No offense RMS. :)

I didn't know about the yeeloong, though. Interesting device.

I also use emacs -nw on my netbook. Its faster than the graphical emacs and, if you turn off the menus, its a quiet minimalist experience.

> I didn't know about the yeeloong, though. Interesting device.

It seems to be very slow, has a small RAM and just a couple gigabytes of storage. I would love the idea of a Windows-proof computer I could use, but this is still not it.

The larger one looks promising.

Which, based on what RMS describes, is perfectly fine for his needs.

Well... It's not for me. I can't recommend it.

"accidentally touching the mouse-pad and moving the pointer, which would be a nuisance"

He should just use a trackball.

It's probably for speed and battery life. Whatever works for you.

Ironic that the only laptop running free software from top to bottom is made in China.

Pretty cool little computers they make too. Check out this: http://www.lemote.com/bencandy.php?fid=134&id=675

RMB¥1699, that's about AUD$295 or USD $265.

Or if you prefer the iMac form factor: http://www.lemote.com/bencandy.php?fid=136&id=849

WTF is with the name of that PHP script though.

Right, apparently free software is more important than actual freedom in Stallman's mind.

Where was your computer built?

Never said I was a Freedom (as in "Freedom") activist like RMS.

Full screen Emacs on a laptop? I wouldn't want a bigger screen than 13' with that setup.

RMS is my favorite computing austerity artist.

The RMS Waltz: That's one, two, three internet connections. Now here's one Emacs (Gnus is not responding), two Emacs (dired and the tramp), three Emacs (at this time, Count von Count can be heard laughing in the background).

Reply on news.ycombinator.com