Fyndo's musings

Can Free/Open Source Software Innovate

noreply@blogger.com (Fyndo) — Sun, 11 May 2008 02:25:00 +0000

The other day in the bookstore I was leafing through a book, and it made the argument that Free/Open Source software was very good at creating (possibly better) copies of existing programs, but wasn't good at breaking new ground.

Personally, I think this argument is stupid. While Tim Berners-Lee's first WWW browser wasn't released under a "free" license, but it wasn't really closed-source either: His announcement said: "If you're interested in using the code, mail me. It's very prototype, but available by anonymous FTP from info.cern.ch. It's copyright CERN but free distribution and use is not normally a problem.", and the (later dominant) NCSA Mosaic was released under an open source/free license. Web servers were similarly free, and honestly, nobody has ever even caught up to Apache and the other free web servers in terms of quality. Netnews was mostly run on free software, as was much of the internet (much of that code from Berkley). Clearly that was innovation.

CVS (the version control system) was open-source, and was incredibly innovative, and there's been incredible amounts of innovation in that realm by open-source projects. Cfengine (the system management tool) was pretty innovative. Perl. Python. Ruby. Rails. PHP. In games we have nethack/rogue etc (which went on to inspire the creation of Diablo), as well as the many MUDs.

So why do people have this absurd notion that Free Software can't innovate? I can think of a couple reasons. The main one is that FOSS development is greatly affected by communication costs. The simpler sharing your source code is, the easier a distributed development model is. Thus, the amount of Free Software written, and the ease of the Open Source Software development model shot up enormously along with the growth of the internet. So anything that pre-dates the internet wasn't going to be developed as Free Software. So yes, Free spreadsheets/WYSIWYG wordprocessors/etc. are fairly derivative. Proprietary software had a substantial head start. But in internet technologies (especially on the server side) Free software has the lead, since you had the internet available to collaborate over if you were writing an internet server.

I suspect proprietary development also allows more rapid development of mid-sized software, so Netscape got all the fame for the Web, because it was able to step in at the right time and make a better browser than Mosaic quickly. But that doesn't mean it's any more innovative.

And, of course, people think it because it's what Microsoft tells them to think ;)

Biomaterials

noreply@blogger.com (Fyndo) — Fri, 28 Mar 2008 05:27:00 +0000

Ok, I used to be a little skeptical of the whole "let's imitate nature to make better materials" thing, then I met someone who'd done work in materials, and worked with those sorts of projects. Now I'm very skeptical. But this article about squid beaks is cool. Although that's probably just because I like squid.

How theorists save lives

noreply@blogger.com (Fyndo) — Wed, 26 Mar 2008 05:06:00 +0000

From xkcd.

See how useful us theorists are?

Or harmless or something.

Arthur C. Clarke Dies.

noreply@blogger.com (Fyndo) — Wed, 19 Mar 2008 18:06:00 +0000

Not that this is news to anyone, but Arthur C. Clarke just passed away. The New York Times obituary is as good as any. Gygax and now Clarke, March is not being kind to geekdom.

If Version Control Systems were Airlines

noreply@blogger.com (Fyndo) — Fri, 07 Mar 2008 03:35:00 +0000

Just saw this article on If Version Control Systems were Airlines, and enjoyed it. Though I will state for the record that I understood Darcs' theory of patches on first reading (perhaps unsurprising as I use the physics theory it's modeled on professionally).

Idiotic stuff about the Patent Reform Act

noreply@blogger.com (Fyndo) — Thu, 06 Mar 2008 15:32:00 +0000

The Patent Reform Act will harm the U.S. technology industryOh great. Stupid merchantilist logic being used to defend the patent system. "It's bad for foreigners, it must be good for us". But then I also buy these products, and lose out if they're more expensive.

And honestly, if some jackass patents the french fry, and sues McDonald's... That's just abuse of the patent system. That lawsuit does nobody (besides the lawyers) any good. When I deliberately don't enforce (or publicize) a patent so that people can get good and dependent on it before suing... how does that do the economy as a whole any good? Transparency is good for the economy. Grrmph.

Physicists Successfully Store and Retrieve Nothing

noreply@blogger.com (Fyndo) — Fri, 29 Feb 2008 23:03:00 +0000

I have to look up a copy of the actual article from Phys Rev Lett, but this is incredibly cool: Physicists Successfully Store and Retrieve Nothing

Basially, it seems to me the researchers managed to create a stationary excited vaccum state. The article describes it about as well as I can (especially since I haven't read the paper yet, and my quantum electrodynamics kung-fu isn't all that strong), so you should probably read that, but... Wow.

Hygienic Macros, Embedded Languages, and Lisps

noreply@blogger.com (Fyndo) — Thu, 28 Feb 2008 07:15:00 +0000

So I've been writing (common) lisp code the last couple weeks in my spare time (or some of my spare time), and was thinking about the language, and realized why I never really liked Scheme's hygienic macros. The macro system is basically it's own little embedded (turing complete, as it turns out) language. So when you decide "hey, I need a macro" you need to mentally switch to a new language. You've been writing various list-manipulation code with first, rest, do, whatever tools you like, and all of a sudden... you're using a new, different language to .... manipulate lists. And it's not a trivial/easy language. Loop is a special little language embedded in Lisp, but pretty straightforward. Who really can't guess what (loop for i from 1 to 10 summing i) does? (Actually, it's a fairly cobol-like language, on reflection, funny given how much I like loop, and mock cobol ;)

Kenny, McCarthy, Lisp, and the Future

noreply@blogger.com (Fyndo) — Wed, 27 Feb 2008 07:58:00 +0000

Kenny Tilton has a cool bit on his blog about why Lisp is so cool: smuglispweeny: Ooh! Ooh! My turn! Why Lisp? Personally, I liked the story at the end:

At ILC 2002 former Lisp giant now Python advocate Peter Norvig was for some reason allowed to give the keynote address like Martin Luther leading Easter Sunday mass at the Vatican and pitching Protestantism because in his talk Peter bravely repeated his claim that Python is a Lisp.
When he finished Peter took questions and to my surprise called first on the rumpled old guy who had wandered in just before the talk began and eased himself into a chair just across the aisle from me and a few rows up.
This guy had wild white hair and a scraggly white beard and looked hopelessly lost as if he had gotten separated from the tour group and wandered in mostly to rest his feet and just a little to see what we were all up to. My first thought was that he would be terribly disappointed by our bizarre topic and my second thought was that he would be about the right age, Stanford is just down the road, I think he is still at Stanford -- could it be?
"Yes, John?" Peter said.
I won't pretend to remember Lisp inventor John McCarthy's exact words which is odd because there were only about ten but he simply asked if Python could gracefully manipulate Python code as data.
"No, John, it can't," said Peter and nothing more, graciously assenting to the professor's critique, and McCarthy said no more though Peter waited a moment to see if he would and in the silence a thousand words were said.

John McCarthy is a really smart man. Personally, I'm also a big fan of his web pages on the sustainability of human progress. Every time some enviro-weenie talks about "sustainability" I want to smack them. Back to Kenny for a moment, at the last Lisp NYC meeting, Kenny was complaining (mentioned, perhaps, lest anyone think Kenny is a whiner) about his shoulder, that he hurt playing Tennis. The enviro-weenie would leap in, say "Tennis is unsustainable, you need to stop playing". Kenny wisely just corrected his form, stopped doing stupid things, and the pain got better. He said something to the effect of "Pain moves you towards correct form". Environmental troubles may suggest better ways of doing things, but there's rarely a need to stop.

Linux Rant

noreply@blogger.com (Fyndo) — Tue, 26 Feb 2008 04:29:00 +0000

Figured since I just got added to Planet Nylug, I'd make a Linux related post. This is going to start out as a Gnome rant, but is much more generally applicable.

I switched from gnome a few months ago when the latest gnome release started running Nautilus on my desktop every time I logged in with no way to turn it off (the left edge of my screen is where the terminal windows go, running some huge bloated file manager just to put some $#%@@$%% icons where I'll never see them is a pain). Finally, I did a web search and found that you had to use gconf to enter an undocumented (as far as I could tell) option to not run it automatically. I'd already gotten sick of having to learn a new undocumented sooper-sekrit gconf option every release to just get gnome-terminal to use the ancient 6x12 bitmapped font (-misc-fixed-medium-r-semicondensed--12-110-75-75-c-60-iso8859-1), and I'd had it. (I have, incidentally finally figured out a solution to getting 6x12 to work easily in KDE (and probably Gnome as well), I copied the font files, changed them from "semicondensed" to "normal", and changed the foundry to "fyndo", and the KDE font picker finds them now that it doesn't need to understand "semicondensed").

Sure, they're trying to make it more user friendly, but "Google the mailing list" is not user-friendly. Control panels can be user friendly, if they're complete. But even if you have an "intuitive" user interface, You need good docs. And Linux docs these days suck. I can almost live with the gnome people taking options off the control panel and making me set them directly with gconf, except they aren't documented. gnome-help has a nice friendly thing about "The Desktop", the nautilus man page documents the --no-desktop option:

--no-desktop
Do not manage the desktop — ignore the preference set in the preferences dialog.

, but since it's being started automatically, that does me no good, and the preference is no longer in the preference dialog. Wrong docs are not good docs. If I have to use gconf to configure something, there should be some reference that lists the gconf options, or they should be self-documenting like emacs. (In emacs functions have doc strings right in the function, and variables have doc strings, so that when you search for the variable to set with "apropos" you can get the docs. Also, since the docs are in the source, if you change a function's behavior, it's obvious where to document it). When I started using Unix back in 1988, I fell in love, because almost everything was configurable. Looking back, the key point was that how to do it was also well documented. I'm still fuzzy on how it is that KDE manages to launch xine when I put in a DVD. Or how to get it to do the right thing. Ok, it used to be the case (yesterday) that I'd put in a DVD, it'd start xine, which would complain it doesn't know what to to do with system:/media/hdc, and I'd have to click on "ok" and then the "DVD" button, and the DVD played. I decided I'd fix that while writing this post, so I went to the control center, and "KDE Components" and "File Associations", and.... it wasn't there. I knew I'd set it to use xine somewhere so I kept looking, and under "Peripherals" and "Storage Media", I found the option for "DVD Video Disk", and changed the command "xine %u" to "xine dvd:/", and now when I put in a DVD, I get a breaking glass sound, a dialog pops up with "/media/cdrom0 is a folder, but a file was expected". Now, the command "xine dvd:/" typed at a command line works ok. There's a little ? icon on the upper right that I take to be some kind of a help icon. Pressing it brings up the incredibly informative "FIXME: Write me" message. There's also a "Help" button on the bottom. It does, as best I can tell, nothing, and I can't find docs for the "peripherals" module. What does that %u do? Beats me! But it's important. If I leave it off, KDE doesn't even try to run the command I have there. I've tried replacing xine with a shellscript running in an xterm, and without the %u it doesn't run. Why? I don't know, the fact that if it DID use that command, it'd work seems irrelevant. And I have no idea how the "a DVD has been inserted" notification gets from the kernel (well, ok, it uses dbus, I think) to KDE. Sure, getting these things to "just work" is complicated, but that's no excuse for not explaining how they work. Neal Stephenson, in his "History of the Command Line" say that when linux fails, it does so transparently. That is no longer true, and I miss it.

Playing Games With Kosovo

noreply@blogger.com (Fyndo) — Mon, 25 Feb 2008 23:05:00 +0000

Wes Clark has a discussion of the current Soviet^H^H^H^H^H^HRussian opposition to the independence of Kosovo. Playing Games With Kosovo | Newsweek International Edition | Newsweek.com. The more tings change, the more they stay the same. Russia really needs to get over it's imperial pretensions so that it can get around to being a great country.

Thoughts on the generational divide in copyright morality

noreply@blogger.com (Fyndo) — Tue, 25 Dec 2007 00:06:00 +0000

I saw a link to this blog entry at the times by David Pogue, and it prompted a whole bunch of thinking on my part, and I figured I'd just make (some of) that thinking available to everyone. The basic premise of the post was that when he gave a talk at a college, none of the students, or no more than a tiny handful, seemed to find the most egregious example of copyright violation immoral.

Reading the comments there were some points made I thought were good:

College students never raise their hands in a lecture. Ever. (I teach college for a living, it's true). So there's some false negatives there.
What's new? I had a substantial collection of cassette tapes in college I'd copied off friends. Traded them when in college. College students, and highschool students are frequently broke. When it's a case of choosing between unpaid for music, and no music... many will opt to not pay for it. If that behavior is endemic amongst the other students, who wants to be the one throwing stones?

There was another point made, that the students "hadn't been educated on the morality of copyright", and that I thought was largely bunk. I know I've seen trailers/previews at the movies trying to convince me not to download movies. I've seen ads against file sharing, and music downlading. And I'm.... less plugged in to such things than most teenagers/college students. The problem is not that nobody has tried, it's that the copyright industry's attempt to educate kids is stupid, and probably morally bankrupt. Copyright is different than theft. When you steal a CD, or DVD from the store, then the store is poorer by one CD or DVD. When you download a movie, or song, then someone may have been deprived of an opportunity to make money to sell it to you, but they are not poorer. That does not mean it's not wrong, but the two situations are not equivalent. And honestly, kids are smart enough to see that. So if you couch the argument (as the copyright industry does) in those terms, you're fighting an uphill battle.

Which brings me to the more general topic of morality and law. But that, I think for another day.