sourcefrog

Keystone of Mormonism

Because of the recent mentions of SCO and Utah on this site, Google inserted an ad for the book The Keystone of Mormonism, “an account of the author's struggle to overcome Mormon indoctrination.”

A couple of chapters are available on the web site and they make interesting reading: both a very personal story, and also a study of techniques of psychological control:

I was taught as a small child that the most important thing for any person to do in this life was to gain a "testimony" of Mormonism and then remain faithful to it forever. The worst sin that a person could commit wasn't lying, stealing, adultery, or even killing. It was apostasy! Other sins could be repented of and forgiven. The Doctrine And Covenants, clearly teaches that there will be no forgiveness in this life nor in the next for a man who receives the Melchizedek (higher) priesthood and then changes his mind and turns against Mormonism.1 Apparently, conversion to Mormonism is supposed to be a one-way street.
My parents did, in fact, belong to the Church. It owned them, and they owned me. I trusted completely in the wisdom and honesty of my parents. It never occurred to me that they could be wrong or even partly wrong or misled. I was quite sure that any negative feelings or thoughts that I might have about the Church were only because I was young, immature, ignorant, weak, or perhaps even tempted by the Devil.

posted Tue 30 Sep 2003 in /issues/religion | link

Really fast

leighklotz writes:

Last weekend I got a call from Comcast offering me Cable Internet service for an introductory rate of $21.95/mo. I asked how fast and the telecaller said, "Six hundred and thirty five gigabytes." I said, "Per month? Per hour? Per second?" She said, "Per second, sir."
I asked, "Can I run servers?"
She said, "Yes sir!"
I said, "On port 80 and port 25?"
She said, "On all ports, sir."
I said, "Before I sign up I'd like to speak to your supervisor to confirm this great deal."
Sadly, the deal evaporated when I got to speak the the sympathiser, but she was interested in what I wanted. I told her I had 1Mb/1Mb symmetric access and static 8 IP addresses, and she asked what they could do to get me to move to Comcast Cable Internet service. I suggested perhaps symmetric service 1.5Mb/1.5Mb would be nice, or perhaps 3Mb down and a portable Class C netblock to do multi-homing with my current 1Mb SDSL uplink. She wrote it all down and said she'd pass my request along.
I'm still smarting at the loss of the 635GB/sec downlink for $21.95/mo though!

posted Tue 30 Sep 2003 in /random/humour | link

A patch

I wrote a patch to do this. Not perfect yet but it works.

For some reason even though many programs want to set the xterm title, the data to do this properly is not in the distributed termcap/terminfo file. Various packages kludge it by e.g. hardcoding the sequences for xterms, or overriding termcaps, as in Debian's /etc/screenrc. I think it might be better just to patch the terminfo database.

posted Mon 29 Sep 2003 in /software/emacs | link

Free Software Kiai

Stephen J Turbull writes:

I think they understand the spirit of cooperation; I much doubt they understand the "kiai" (unyielding martial spirit) of Copyleft. Copyleft is a double-edged sword; it can inhibit, as well as promote, cooperation.

posted Mon 29 Sep 2003 in /random | link

Really, I'm not out to destroy Microsoft. That will just be a completely unintentional side effect.
— Linus

posted Mon 29 Sep 2003 in /random | link

BMW K1200 GT review

And now for something completely different...

I also had the pleasure of riding a BMW K 1200 GT, on evaluation from Rolfe Classic in Canberra.

(As an aside: why would you have a motorcycle salesperson who doesn't ride, and therefore doesn't really have any vocabulary in common with their potential customers? I suppose if people walk in knowing exactly what they want, anyone can negotiate price, but sometimes they want some information or advice too... All the answers about how it compares to other models, or what it's like to live with for more than an hour can only come from experience.)

Back to the bike. The executive summary would have to be beautiful bike; shame about the motor.

To get the worst out of the way: the 4 cylinder 1171cc motor isn't really bad; it's just a bit bland. It produces usable torque from a touch above idle all the way up. It's very very smooth; since a sticker obscured part of the tacho on the bike I rode I once wondered if it had stopped. Let the clutch out in first or second or third, smoothly take off, pop up a gear or two. I wonder if the self-effacing engine is one reason why the GT has an LCD gear indicator on the console, something I haven't seen on anything else in years.

In a bike in this class you're not wanting wheelies or burnouts. You don't want your pillion falling off. But corporal engagement with the motor is part of the experience of riding. If you pay AUD 28,000 for a bike it ought to have a motor that's interesting to ride. I think my disappointment is twofold: it accelerates adequately, but unremarkably. It's very noticeably slower than your average 600cc bike, though much of this is likely due to the substantial >300kg mass. Secondly, the power's delivered in such a quiet and smooth way that opening the throttle doesn't bring the thrill we normally enjoy. The torque curve seems pretty flat; there is no top end surge as such.

Perhaps I'm just too young.

Leaving the lack of power aside, it is a wonderful bike in many other respects. It's hard to describe how the mass disappears once it gets moving, but it really does. It's amazing: parked it looks like a lot of frontal area and a lot of weight, but once you're rolling it feels light and well composed. I rode it on a windy day and was worried that all the plastic would catch the sidewinds, but in fact it does very well compared to the smaller ZX12-R. It must be something about the shape, or perhaps the bike's inertia keeping it steady.

Front wind protection is also very good. The screen is electrically adjustable over a small range. The air behind it is extraordinarily calm even at 130+. I felt distinctly warm on a spring day which might have been nippy on less protected bikes. The wind protection plus a very neutral seating position makes it very comfortable.

The switchgear is pretty loaded with controls for heating, the windscreen, cruise control, headlights, hazard lights, and the BMW-style indicators. It takes a bit of getting used to, but your thumbs will be busy at first. I can imagine the BMW indicator setup being easier with winter gloves, which is perhaps why they stick with it.

Heated grips and seats are standard, which I'm sure would be luxury in winter. Panniers are also standard.

The Telelever suspension is strangely disturbing compared to conventional bikes when you first corner or brake. I think the big difference is that it doesn't dive or lift in the same way that conventional forks do; at first you miss the feedback but it soon starts to feel really good and very stable.

The linked/ABS brakes are a tower of strength. They're as good as any OEM sportsbike brake I've seen, even though they need to disipate perhaps twice as much kinetic energy. Press the lever and it stops very rapidly but controllably, and without perceptibly diving. The sharpness and tight control immediately reminded me of braided steel lines I had on a CBR600 a few years ago. It feels enormously safe and probably contributes to feeling easily in control of the heavy unit.

The burning question for observers of BMW motorcycles has to be: why can't the company that makes the M3 and M5 coupes produce a bike to match? We don't need another liter superbike, but something that is luxurious but powerful would be worth seeing.

Lovely bike, but not for me for a few years at least. If you had the money and wanted to do a long highway trip at roughly legal speeds the K 1200 GT would have to be a top contender.

posted Sat 27 Sep 2003 in /motorbikes/review | link

Kawasaki ER-5 review

I had a Kawasaki ER-5 loaner from the good blokes at Canberra Motorcycle Centre for the last couple of days.

What we have here is a nice, competent 500cc twin street bike. For me at roughly 182cm the size was pretty good; it felt small and light but not tiny. The seat is not high; for somebody shorter but with reasonable lower body strength it would probably also be cosy.

The engine note is lovely and burbling; it starts easily from cold. Acceleration from the lights is good, and it has a broader torque band (~2500-9000rpm) than comparable smaller bikes. The relatively large displacement for a learner bike and a heavy flywheel(?) should make it patient with on riders who are still getting the hang of the clutch.

Cornering and braking are pretty nice for what is not really a performance bike. You can scoot around inner city corners with confidence in the slightly narrow tyres (110/70-17R front, 130/70-17R rear) and handling. It's not quite “on rails”, but it is wearing good grippy sneakers.

Steering is of course very direct and tight compared to a sports bike, and the relatively low gearing means that putt-putting around carparks or the inner city or parking on sloping footpath is a piece of cake. It feels rather less than its dry 179kg.

The obvious weak point is the brakes. The single disc front and drum rear both produce nowhere near the force they ought to — and brakes are one thing that's not really optional. A very tight squeeze on the front produces rather asthmatic deceleration, and the rear brake seems to have less effect than engine braking.

I suppose the nicest thing I can say about the brakes is that they might be comfortable for a novice who's afraid of sliding the front wheel under heavy braking. However hard I squeezed, it didn't seem to get anywhere near the friction limit of the modest front tyre. The price of unthreatening braking might well be sliding right into a crash that you could have avoided on a small sports bike.

So the ER-5 is roughly AUD $8900 brand spanking new, friendly to ride, and very practical around town. Under the current laws in the ACT and most of Australia, it's learner-legal because of the conservative power-to-weight ratio. Not only is the ratio legal, but I think the absolute values are sane: it's not so heavy as to get away from you, and it's unlikely to lead the new rider into too much temptation.

I would almost like to say it'd be a good learner bike, but the brakes worry me. I hate the idea of a learner drifting into a collision with the calipers squeezed shut. Perhaps I'm being unfair here — I haven't sampled the competition recently. But from memory even something like a Balius feels far safer. Before buying an ER-5 I'd definitely make sure to pay attention to the brakes. Aside from that, go ahead and enjoy.

posted Sat 27 Sep 2003 in /motorbikes/review | link

Snopes on darkprofits

The ever-reliable Snopes urban legends site reports on darkprofits. This seems to be one of a stream of joe-job attacks on them. The site seems to be down now, perhaps as a result. A page that used to be on darkprofits says:

Please note that we do not send Spam of any kind. The Spam you received is from a disgruntled banned member of our forum. We thank our host for seeing that the claims of content are ludicrous and this is what is known as a "joe job".
We apologize for the Spammer's actions and if you are here looking for what is contained in the Spam you will be disappointed. We do not have any porn for you perverts. We also do not have any nuclear weapons or pounds of drugs for you junkies. We are also fresh out of gay slaves, hookers and the like.
Please check the headers of your Spam and complain to the true senders account.

posted Sat 27 Sep 2003 in /issues/spam | link

console-mode frame title sets Xterm title

Here's a feature I'd like to see: console-mode frame title sets Xterm title. I can't find a record of anyone actually doing it yet.

Noah Friedman has xterm-frobs.el which provide a low-level interface but they are not automatically called when the title needs to be updated. It points out the right (?) way to send the escape sequences out: send-string-to-terminal. Perhaps they could be called that way. Noah's code also includes a special case to pass the string through screen to get it to the real terminal, but that's not what I want. I think getting the screen name set properly could be pretty useful.

I wonder if the escape sequences can really be hardcoded in this way or if they have to be fetched from termcap/terminfo? Different terminals might conceivably support it in different ways, though I think the most important ones these days will be close to xterms.

posted Fri 26 Sep 2003 in /software/emacs | link

More on darkprofits

More information on darkprofits from aota.

posted Fri 26 Sep 2003 in /issues/spam | link

darkprofits.com credit card scam

Spam to a mailing list today, trapped by spamassassin:

I wonder what kind of sucker rate they get?

posted Wed 24 Sep 2003 in /issues/spam | link

distcc 2.11 out

I released distcc 2.11, with the new GNOME monitor. Implementing this has been an interesting trip through the GNOME API.

There is semi-infinite scope for adding additional eye-candy, but I think it gives a pretty good overview of what is happening on the system as it is.

posted Tue 23 Sep 2003 in /projects/distcc | link

Diamond Wiki

Diamond Wiki is an attempt to combine faceted navigation with wiki- and blog-style features. Diamond Wiki is written by Kim Burchett, based on my PikiPiki simple Python wiki.

Kim's blog has other interesting thoughts about metadata and categorization.

posted Mon 22 Sep 2003 in /projects/piki | link

gtk_widget_hide_on_delete

To make a GNOME modeless dialog just hide when its frame Close button is clicked, connect delete_event to gtk_widget_hide_on_delete. The default is to destroy the widget, which can be inconvenient if, for example, it was constructed by Glade and you don't want to rebuild it when the user wants to see it again.

posted Mon 22 Sep 2003 in /software/gnome | link

memcached: a distributed memory object caching system

memcached is a high-performance, distributed memory object caching system, generic in nature, but intended for use in speeding up dynamic web applications by alleviating database load.
Danga Interactive developed memcached to enhance the speed of LiveJournal.com, a site which was already doing 20 million+ dynamic page views per day for 1 million users with a bunch of webservers and a bunch of database servers. memcached dropped the database load to almost nothing, yielding faster page load times for users, better resource utilization, and faster access to the databases on a memcache miss.

posted Mon 22 Sep 2003 in /software | link

Colin Walters on Arch

Colin writes approvingly of Arch.

I recently switched to using the Arch revision control system for Rhythmbox, and I am incredibly happy with it now. I've already had one person who had no account on the GNOME CVS make a local branch of my tree (without write access to it either), start hacking on it, and then I was able to star-merge their changes back in easily. We've actually merged since several times as he made more bugfixes. There have been no conflicts from repeatedly applied patches. Great stuff.

It looks good to me. I just wish there were more documentation, but I suppose that will come in time. In the meantime, he has an IRC log with some description of how to use it.

posted Mon 22 Sep 2003 in /software/vc/arch | link

Bitshifts in C

What do you think this does?

uint32_t a, b;
[....]

b = 32;
a = a >> b;

You might think this will reduce a down to zero. But in fact, the C99 specification says that shifting either left or right by more than the width of the type causes undefined behaviour. On gcc on i386, it in fact shifts by b % 32, so in this case a is unchanged.

posted Mon 22 Sep 2003 in /software/languages/C | link

Sun continue to amaze

I don't normally write about this kind of stuff, but Jonathan Schwartz's eWeek interview is just too good:

Also, let me really clear about our Linux strategy. We don't have one. We don't at all. We do not believe that Linux plays a role on the server. Period.

Wow, talk about attention-grabbing quotes. Even Microsoft have a Linux strategy these days!

Further wisdom on this topic is available on Slashdot.

posted Mon 22 Sep 2003 in /software/industry | link

Microsoft blog complaining about IIS is down...

I see that gazitt.com is currently not responding. This was the site that mentioned the way IIS is intentionally crippled to drive demand for more expensive systems. I don't know that IIS is the problem — it could be some kind of hardware failure, or perhaps Windows crashed — but it is poetic.

posted Mon 22 Sep 2003 in /meta | link

Trying it out

OK, I have nxml loaded, and I'm trying it out for the slightly incorrect HTML fragment mode that Blosxom uses. So far so good.

One nice thing compared to psgml mode is that complains a little less about documents that aren't quite proper XML. It is actually useful to edit things that are incomplete or incorrect but psgml whines so much it's not worth trying.

posted Sun 21 Sep 2003 in /software/emacs | link

Microsoft Blogs

A few people have commented on the recent emergence of blogs written by Microsoft people. I think this one is a particularly nice example. He's talking about somebody filing a bug complaining about the milk in the cafeteria, which is something that might happen in many development teams. But there's something about the style of writing that is particularly Microsoftie. The explanation could have come straight from The Road Ahead:

This is a particularly software-oriented joke, because it highlights how hard it is to make bugfixes in software - by applying the software testing regimen to something that isn't software. You can't assume that a simple, local change like adjusting the amount of glue applied to the carton will result in a simple, local change in the final product (a more acceptable seal strength). Software is nonlinear. A simple change can have effects (some catastrophic, some subtle) far, far away from the point of change.

Further exhibits: simplegeek.com, winethirty.com.

Following on from there, autocode writes about syntax highlighting and related editor things in Visual Studio. I was just writing about emacs, so it's kind of interesting. And in a neat circularity, prolific Microsoft/XML author Don Box knows emacs is the finest editor.

Yet further on, somebody is complaining about performance problems with his blog. Certainly a common enough topic, since blogging is probably bringing more people into running little active web applications than ever before. But here it turns out that the problem is in fact that Windows XP's web server is limited to only 10 HTTP connections, even on hardware that could probably serve thousands. This is for "business model reasons" — Microsoft wants people to run a more expensive system for web servers. There may be a time and a place for proprietary licensing, but intentionally crippling software so that your customer's customers see error messages just seems very poor.

posted Sun 21 Sep 2003 in /meta | link

The Art of Unix Programming released

TAOUP has gone to press. I think it will be worth buying; for more review notes see /weblog/books/taoup/.

posted Sun 21 Sep 2003 in /books/taoup | link

James Clark unveils a new XML mode for GNU Emacs

xmlhack:

More magic from James Clark: He's announced the alpha release of nXML, a new mode for editing XML documents from within GNU Emacs. It's a milestone in that it's the first open-source editing application to enable context-sensitive validated editing against Relax NG schemas. It also provides a clever mechanism for real-time, automatic visual identification of validity errors, along with flexible syntax-highlighting capabilities -- and many other features planned for future releases.

posted Sat 20 Sep 2003 in /software/emacs | link

Comprehensive SCO fact set

PJ and the Groklaw community have written a superb comprehensive summary of the SCO suit for The Inquirer.

Linus Torvalds on his attitude toward IP: "Torvalds took issue with SCO's position. 'I care deeply about IP (intellectual property) rights. I've personally got more IP rights than the average bear, and as the owner of the copyright in the collective of the Linux kernel, I shepherd even more. It's what I do, every day. I personally manage more valuable IP rights than SCO has ever held, and I take it damn seriously,' Torvalds said in an e-mail interview."

posted Sat 20 Sep 2003 in /issues/sco-vs-linux | link

Progress on distcc monitor

Getting something that looks good, conveys the right information, and does not burn an unreasonable number of cycles is a bit tricky.

GTK+ is very nice to program too. C is not ideal for this kind of stuff though.

posted Thu 18 Sep 2003 in /projects/distcc | link

If airlines sold paint...

(Forwarded):

Customer: Hi. How much is your paint?
Clerk: Well, sir, that all depends on quite a lot of things.
Customer: Can you give me a guess? Is there an average price?
Clerk: Our lowest price is $12 a gallon, and we have 60 different prices up to $200 a gallon.
Customer: What's the difference in the paint?
Clerk: Oh, there isn't any difference; it's all the same paint.
Customer: Well, then I'd like some of that $12 paint.
Clerk: When do you intend to use the paint?
Customer: I want to paint tomorrow. It's my day off.
Clerk: Sir, the paint for tomorrow is the $200 paint.
Customer: When would I have to paint to get the $12 paint?
Clerk: You would have to start very late at night in about 3 weeks. But you will have to agree to start painting before Friday of that week and continue painting until at least Sunday.
Customer: You've got to be *&%#@* kidding!
Clerk: I'll check and see if we have any paint available.
Customer: You have shelves FULL of paint! I can see it!
Clerk: But it doesn't mean that we have paint available. We sell only a certain number of gallons on any given weekend. Oh, and by the way, the price per gallon just went to $16. We don't have any more $12 paint.
Customer: The price went up as we were talking?
Clerk: Yes, sir. We change the prices and rules hundreds of times a day, and since you haven't actually walked out of the store with your paint yet, we just decided to change. I suggest you purchase your paint as soon as possible. How many gallons do you want?
Customer: Well, maybe five gallons. Make that six, so I'll have enough.
Clerk: Oh no, sir, you can't do that. If you buy paint and don't use it, there are penalties and possible confiscation of the paint you already have.
Customer: WHAT?
Clerk: We can sell enough paint to do your kitchen, bathroom, hall and north bedroom, but if you stop painting before you do the bedroom, you will lose your remaining gallons of paint.
Customer: What does it matter whether I use all the paint? I already paid you for it!
Clerk: We make plans based upon the idea that all our paint is used, every drop. If you don't, it causes us all sorts of problems.
Customer: This is crazy!! I suppose something terrible happens if I don't keep painting until after Saturday night!
Clerk: Oh yes! Every gallon you bought automatically becomes the $200 paint.
Customer: But what are all these, "Paint on sale from $10 a liter" signs?
Clerk: Well that's for our budget paint. It only comes in half-gallons. One $5 half-gallon will do half a room. The second half-gallon to complete the room is $20. None of the cans have labels, some are empty and there are no refunds, even on the empty cans.
Customer: To hell with this! I'll buy what I need somewhere else!
Clerk: I don't think so, sir. You may be able to buy paint for your bathroom and bedrooms, and your kitchen and dining room from someone else, but you won't be able to paint your connecting hall and stairway from anyone but us. And I should point out, sir, that if you paint in only one direction, it will be $300 a gallon.
Customer: I thought your most expensive paint was $200!
Clerk: That's if you paint around the room to the point at which you started. A hallway is different.
Customer: And if I buy $200 paint for the hall, but only paint in one direction, you'll confiscate the remaining paint.
Clerk: No, we'll charge you an extra use fee plus the difference on your next gallon of paint. But I believe you're getting it now, sir.
Customer: You're insane!
Clerk: Thanks for painting with United.

posted Wed 17 Sep 2003 in /random/humour | link

wdired mode

A little-known (to me :-) gem of emacs is wdired-mode, which is in emacs-goodies-el on Debian. With this installed, you can press r in the directory editor, and change the names of files by directly writing over them, more or less as in Nautilus or Mac OS.

The Angry Moron Newsletter

Tim has a new web page, and it reminded me of the Angry Moron Newsletter, to which I was briefly and involuntarily subscribed. It's even funnier if you imagine Mr Gumby doing all the voices, and them all talking over each other.

posted Wed 17 Sep 2003 in /random | link

Multics Emacs

Following on from an earlier post on the history of vi, Luke Gorrie points to Bernard S. Greenberg's Multics Emacs paper:

Wonderful stuff. You just don't get the fun of having to move parts of your input/redisplay code into device drivers today.

I was saying the other day how disappointing it is to me that university Human-Computer Interaction courses always lionize a particular set of projects and in particular the Xerox Star. While the Star was quite an achievement, I think it has become a bit hackneyed as an example. There are plenty of other designs which in their historical context were very innovative and which teach equally or more important lessons. Moving from line editors to interaction may be one.

posted Wed 17 Sep 2003 in /software/emacs | link

Fort Collins

Fort Collins CO

posted Wed 17 Sep 2003 in /random | link

Display filters in the the GIMP

Have you ever wondered what it's like to be color blind? GIMP now has Display filters that change the way the image is displayed without affecting the underlying data. (Image/View/Display Filters...) This could be useful for people designing accessible media.

posted Tue 16 Sep 2003 in /software/gimp | link

Coming soon

Review of Artespresso, Canberra
ZX12-R pr0n
Notes on Unison and screen and C-x z
...

posted Mon 15 Sep 2003 in /meta | link

Bill Joy on writing vi

The Register quotes a Linux Mag interview with Bill Joy on the design constraints for writing vi:

What happened is that Ken Thompson came to Berkeley and brought this broken Pascal system, and we got this summer job to fix it. While we were fixing it, we got frustrated with the editor we were using which was named ed. ed is certainly frustrating.
We got this code from a guy named George Coulouris at University College in London* called em - Editor for Mortals - since only immortals could use ed to do anything. By the way, before that summer, we could only type in uppercase. That summer we got lowercase ROMs for our terminals. It was really exciting to finally use lowercase.
So we modified em and created en. I don't know if there was an eo or an ep but finally there was ex. [laughter] I remember en but I don't know how it got to ex. So I had a terminal at home and a 300 baud modem so the cursor could move around and I just stayed up all night for a few months and wrote vi.

Linux Mag then asked: "So you didn't really write vi in one weekend like everybody says?"

No. It took a long time. It was really hard to do because you've got to remember that I was trying to make it usable over a 300 baud modem. That's also the reason you have all these funny commands. It just barely worked to use a screen editor over a modem. It was just barely fast enough. A 1200 baud modem was an upgrade. 1200 baud now is pretty slow.
9600 baud is faster than you can read. 1200 baud is way slower. So the editor was optimized so that you could edit and feel productive when it was painting slower than you could think. Now that computers are so much faster than you can think, nobody understands this anymore.
The people doing Emacs were sitting in labs at MIT with what were essentially fibre-channel links to the host, in contemporary terms. They were working on a PDP-10, which was a huge machine by comparison, with infinitely fast screens.
So they could have funny commands with the screen shimmering and all that, and meanwhile, I'm sitting at home in sort of World War II surplus housing at Berkeley with a modem and a terminal that can just barely get the cursor off the bottom line.
It was a world that is now extinct. People don't know that vi was written for a world that doesn't exist anymore - unless you decide to get a satellite phone and use it to connect to the Net at 2400 baud, in which case you'll realize that the Net is not usable at 2400 baud. It used to be perfectly usable at 1200 baud. But these days you can't use the Web at 2400 baud because the ads are 24KB.

Even now latency from (say) Australia to the US is high enough that being able to type a long way ahead in the way that vi supports is quite worthwhile.

George Coulouris writes, about vi:

The system support person (I don't recall his name, but I have a vague recollection that he went to Bell Labs later. Bill Joy might be able to tell you his name) said something like: "That's very nice, but if we made it available to all of our users the overheads associated with running in raw mode (a process swap on each key depression) would swamp the cpu".
I was rather depressed by this reaction, thinking "I guess I have been unrealistic in developing an editor that is so expensive to run - it's ok in our small system, but it's no use in the big Unix environment at Berkeley".[...]
By the way, 'em' stands for 'editor for mortals' - I christened it that after Ken Thompson visited our lab at QMC while I was developing it and said something like: "yeah, I've seen editors like that, but I don't feel a need for them, I don't want to see the state of the file when I'm editing".

posted Mon 15 Sep 2003 in /software/people | link

The View from Utah

Off GROKLAW's forum: mobucote talks about embarrasment that McBride is causing to Mormons:

When a fellow Mormon is mentioned in the media I usually feel excitement for the accomplishments of that person. However Darl McBride's behavior is hardly something to feel pride over. I feel his business ethics are questionable and embarrassing to his religious community. I hope no further reference in the media will be made to Darl McBride and his religion for the sake of all Mormons.

I think Mobucote (whose MT "Reply" link is broken) is a little oversensitive. "$company_name, a $location based $category company today announced" is absolutely standard cliched boilerplate for the start of a business news story. A fair fraction of the stories mention that IBM's headquarters are in Armonk NY, as if that has anything much to do with their global operations.

Aside from that - yes, Utah is a slightly unusual location for a software company (i.e. it's not California). No more strange than mentioning "Ottawa", or "Helsinki" or "Sydney". Yes, I realize Novell and WordPerfect are in Utah. Making the connection to them is one reason why it's interesting to mention the location.

Another reason is the presence of the Canopy Group in many Utah technology companies. Apparently they have a deeper involvement in this shambles than merely being a SCO shareholder. IBM's subpoena of Canopy may bring out more details.

Why does the Economist and other publications mention McBride's Mormon belief? Partly just human color for the story. Partly because if you publicly make some kind of commitment to a particular moral code, people are going to expect you to live up to it on little matters like honesty and fair dealing.

posted Mon 15 Sep 2003 in /issues/sco-vs-linux | link

Musings on LCA Papers

I'm reading through the paper abstracts submitted to LCA. This is really hard: there are so many good papers and only a limited number of slots.

The breadth of the submissions makes me happy too. We have technical breadth (sound recording to databases) and depth (Linux in schools to kernel hackery). There are submissions from people on six continents, and in particular more interest from Asia than in previous years. LCA seems to be gaining quite a reputation.

Making decisions about which papers will work well implies some idea of what the focus of the conference ought to be. We probably have enough submissions to run an Australian Kernel Summit if we wanted, and I for one would go to every session. But I don't think that best advances our World Domination mission: we need topics that will be approachable and exciting for people relatively new to Linux, and even old farts can benefit from looking up from their keyboard to see what else is happening.

I'd like people to go home from LCA feeling excited about what's happening with Linux and free software and motivated to deepen their involvement. Perhaps they'll install some new bit of software they heard about; perhaps they'll try modifying their kernel; perhaps they'll install Linux at their work.

posted Mon 15 Sep 2003 in /projects/lca2004 | link

Which Doctor are you?

You are the Fourth Doctor: A walking Bohemian
conundrum with a brooding personal magnetism
and a first-rate intellect concealed somewhere
beneath your charmingly goofy exterior. You are
perhaps the most terribly clever of all the
Doctors, though your occasional bouts of
childishness get you in trouble. You never go
looking for a fight, but when someone messes
with you... good heavens, are they ever sorry
they did.

Which Incarnation of the Doctor Are You?
brought to you by Quizilla

posted Sun 14 Sep 2003 in /random | link

JetFile: A Multicast-based Distributed Filesystem

posted Sun 14 Sep 2003 in /software/filesystems | link

Woman and Brutalist building

Woman and brutalist building. Canberra, July 2003, Martin Pool.

posted Sat 13 Sep 2003 in /photo | link

Beginners mistakes in Python

Hans Nowak wrote a good short article on beginners mistakes when moving to Python from some other language.

To preserve cosmic balance, Richard presents anti-pitfalls: mistakes that you can't make in Python that are possible in other languages.

posted Fri 12 Sep 2003 in /software/languages/python | link

Linus's reply to SCO

LWN reports that Linus wrote:

Dear Darl,

Thank you so much for your letter.

We are happy that you agree that customers need to know that Open Source is legal and stable, and we heartily agree with that sentence of your letter. The others don't seem to make as much sense, but we find the dialogue refreshing.

However, we have to sadly decline taking business model advice from a company that seems to have squandered all its money (that it made off a Linux IPO, I might add, since there's a nice bit of irony there), and now seems to play the U.S. legal system as a lottery. We in the Open Source group continue to believe in technology as a way of driving customer interest and demand.

Also, we find your references to a negotiating table somewhat confusing, since there doesn't seem to be anything to negotiate about. SCO has yet to show any infringing IP in the Open Source domain, but we wait with bated breath for when you will actually care to inform us about what you are blathering about.

All of our source code is out in the open, and we welcome you to point to any particular piece you might disagree with.

Until then, please accept our gratitude for your submission,

Yours truly,

Linus Torvalds

Many people have spent hours and pages going through Darl's letter paragraph by paragraph, pointing out inaccuracies, misquotes, misdirection and duplicity. There's plenty of material to work with.

I think Linus has a real talent for detecting bullshit, both in software design and in business letters, and then the good sense to simply refuse it, rather than worrying about the details.

posted Fri 12 Sep 2003 in /issues/sco-vs-linux | link

The Remembrance Agent

The Remembrance Agent (Remem) is an Emacs plug-in that watches over your shoulder and suggests information relevant to what you're reading or writing. While search engines help with direct recall, Remem is a tool for associative memory. Suggested documents are displayed in a buffer at the bottom of your Emacs window, and are updated every few seconds based on the last hundred or so words surrounding the cursor. Documents are pulled from your own text documents, and Remem's internal indexer can parse email archives, HTML, LaTex and plain-text documents.
IBM Systems Journal paper.

I realize this is terribly complacent of me, but it's an OK heuristic: if it's not in Debian, it's probably not worth installing.

posted Wed 10 Sep 2003 in /software/emacs | link

SCO may not know origin of code - Warren Toomey

Sam Varghese reports:

Dr Warren Toomey, now a computer science lecturer at Bond University, said today: "I'd like to point out that SCO (the present SCO Group) probably doesn't have an idea where they got much of their code. The fact that I had to send SCO (the Santa Cruz Organisation or the old SCO) everything up to and including Sys III says an awful lot."
He said that even though SCO owned the copyright on Sys III, a few years ago it did not have a copy of the source code. "I was dealing with one of their people at the time, trying to get some code released under a reasonable licence. I sent them the code as a gesture because I knew they did not have a copy," he said with a chuckle. [...]
SCO was unaware of the origins of much of the code and this "explains how they could wheel out the old malloc() code and the BPF (Berkeley Packet Filter) code, not realising that both were now under BSD licences - and in fact they hadn't even written the BPF code," Dr Toomey said.

posted Wed 10 Sep 2003 in /issues/sco-vs-linux | link

GTK+/GNOME Application Development

Havoc Pennington's book GTK+/GNOME Application Development is now online. It's a good introduction to the interface and covers a lot of the tricks that are not obvious from the API documentation.

posted Wed 10 Sep 2003 in /software/gnome | link

Yussuf al-Ayyeri

Instapundit points to a NY Post column about the book The Future of Iraq and The Arabian Peninsula After The Fall of Baghdad by Yussuf al-Ayyeri.

Al-Ayyeri argues that the history of mankind is the story of "perpetual war between belief and unbelief." Over the millennia, both have appeared in different guises. As far as belief is concerned, the absolutely final version is represented by Islam, which "annuls all other religions and creeds." Thus, Muslims can have only one goal: converting all humanity to Islam and "effacing the final traces of all other religions, creeds and ideologies."[...]
What Al-Ayyeri sees now is a "clean battlefield" in which Islam faces a new form of unbelief. This, he labels "secularist democracy." This threat is "far more dangerous to Islam" than all its predecessors combined. The reasons, he explains in a whole chapter, must be sought in democracy's "seductive capacities."[...]
If established in any Muslim country for a reasonably long time, democracy could lead to economic prosperity, which, in turn, would make Muslims "reluctant to die in martyrdom" in defense of their faith.

posted Wed 10 Sep 2003 in /issues/politics | link

Million lines claim reiterated

Groklaw reports on McBride's recent open letter.

It is interesting to see the "million lines" claim restated, and in writing: SCO claims “that more than one million lines of Unix System V protected code have been contributed to Linux through this model.” I wonder where they are?

posted Wed 10 Sep 2003 in /issues/sco-vs-linux | link

Good rides around Canberra

From the aus.motorcycles FAQ:

Where should I ride in the Australian Capital Territory?
The Uriarra/Cotter loop is a fantastic stretch of road, with wide variations in road 'style' and landscape type. Start on the Cotter road, turn off at Coppins Crossing, then head onto Uriarra road. Follow Uriarra until the Turn-off back to canberra, and it will take you back onto the Cotter road just before Cotter Pub. Continue along Cotter road to Tuggeranong, and return to canberra along any viable route.
You can continue from the Cotter Pub south down past Murray's Corner and Tidbinbilla to Tharwa and back up the Monaro Hwy. The bit through the pine forest past Murray's Corner is a bit prone to slow/unpredicatble traffic and gravel on the road but once it opens up, its quite good, and makes the ride about twice as long.
A quick fang up Coppins Crossing is always fun.
Heading down the coast via the Clyde mountains is an interesting ride with lots of twisties in the later sections, and the ride up the coast from Batemans bay to Sydney (through the national park south of sydney) is a great alternative to the highway (if you have a few extra hours to spare)
The road down Brown Mtn (between Cooma and the coast) is a vastly better road than Clyde Mtn. Every time I went down Clyde Mtn (admittedly a long time ago), the road surface was shocking. Of course, both Brown Mtn and Clyde Mtn are in NSW, not ACT.

I'd like to get photos to do an illustrated version sometime.

posted Tue 9 Sep 2003 in /motorbikes | link

Frank's page

Frank Sorenson has another SCO resource page, including a link to a SCO Princess Bride filk on slashdot.

posted Tue 9 Sep 2003 in /issues/sco-vs-linux | link

IBM Subpoena to Canopy

Groklaw reports that IBM has served a broad subpoena on the Canopy Group, SCO's parent company.

The subpoena includes a definition of the word “including”. Lest there be any doubt.

Also from Groklaw, and on a lighter note, Microsoft have chosen the acronym RMS for their Rights Management Server product, allowing such little gems as

To deploy RMS, organizations are required to have a Windows Server 2003 Server and Client Access Licenses as well as Windows Rights Management Services Client Access Licenses.[...]
An optional RMS External Connector License, which covers an unlimited number of outside users, costs $18,066.

Just imagine, Richard has probably been making external connections all these years without having a proper Licence...

posted Tue 9 Sep 2003 in /issues/sco-vs-linux | link

Taking Mexico by Storm

Chris MacAskill writes about riding San Francisco to Acapulco

posted Tue 9 Sep 2003 in /motorbikes | link

SWEBOK: Sink the Slipper

Cem Kaner tears apart SWEBOK, the so-called “Software Engineering Body of Knowledge”.

SWEBOK was created using a strange process. They started with the table of contents of the main software engineering textbooks -- as if there is a strong relationship between software engineering as described in textbooks and software engineering as practiced in the field. From there, SWEBOK developed as delta's from these books. SWEBOK is focused on "established traditional practices recommended by many organizations" and is intended to exclude "practices used only for certain types of software" and to exclude "innovative practices tested and used by some organizations and concepts still being developed and testing in research organizations."
Somehow, we conclude that mutation testing is an established traditional practice that is widely recommended and used, but we exclude scenario testing. We conclude that massive tombs of test documentation are an established traditional practice widely followed, even though rants about bad test documentation are, to say the least, a common theme of comment in the community. And we exclude consideration of requirements analytical techniques (or project context considerations) that might help you make a sensible engineering determination of what types of documentation, at what level of depth, for what target reader, are worth the expense of creating and (possibly) maintaining them.
In the SWEBOK, page IX, we learn that the purpose of SWEBOK is to provide a "consensually-validated characterization." In this, SWEBOK has failed utterly. Only a few people (about 500) were involved in the project. It alienated leading people, such as Grady Booch who recently said (in a post to the extremeprogramming listserv on yahoogroups, dated 5/31/2003)
"I was one of those 500 earlier reviewers - and my comments were entirely negative. The SWEBOK I reviewed was well-intentioned but misguided, naive, incoherent, and just flat wrong in so many dimensions."

[...]SWEBOK says "Clearly, unit testing starts after coding is quite mature, for instance after a clean compile."
This is 100% in disagreement with the practice of test-driven development, which requires the programmer to write a unit test immediately _before_ writing code that will enable the program to pass the test.
I think that test-driven development is the most important advance in the craft of testing of the past 30 years. This, more than any of the other flaws, illustrates the extent to which SWEBOK is blind to modern good practice.

Why does the criticisim cheer me? Because I know there's a certain type of person who laps up this “pious-sounding claptrap, religious doctrine rather than engineering”. Given half a chance they'll use SWBOK to make their students' or engineers' lives a misery...

posted Fri 5 Sep 2003 in /software/theory | link

LWN coverage

Just after I wrote, I realized that Linux Weekly News also has an article about Conglomerate, which is more detailed. An LWN subscription is excellent value.

posted Fri 5 Sep 2003 in /software/gnome/conglomerate | link

Conglomerate

Alexander pointed out Conglomerate, a very promising GNOME XML structured document editor. The screenshot speaks for itself here:

posted Fri 5 Sep 2003 in /software/gnome/conglomerate | link

Changes between 2.2 and 2.4

SCO have said that there are over a million lines of code on which SCO owns the copyrights in the Linux kernel.

SCO have also said that the 2.2 kernel does not infringe, but all versions of the 2.4 kernel do. [Does anyone have a direct link for this?]

As background: 2.2 is what everybody used a few years ago; 2.4 is what most people use today except for hardy souls trying the 2.6 prereleases.

SCO have not deigned to identify any particular code, but we can do some calcualations just from this information.

What SCO said does not necessarily imply that they think there are a million lines of infringing in 2.4.0; possibly they leaked in over the course of the 2.4 development.

Therefore we can narrow down the potentially infringing code by looking at the changes between the last version of 2.2 (2.2.21), and the version of 2.4 that was current when SCO started this (2.4.20).

I took a unified diff between these two versions. The diff itself is quite hefty at 132MB, but that includes a lot of context lines that were not changed.

diffstat reports the changes are:

 11184 files changed, 3032027 insertions(+), 884233 deletions(-)

Those numbers are in lines, so there were about three million new or changed lines in 2.4.20 relative to 2.2. At this level, it is not mathematically impossible that there were a million lines of SCO code in there.

However, as I previously discussed, I think you need to discount the sections that don't correspond to any existing SCO code. For example, this patch includes these large additions that are not in SCO code to the best of my knowledge. (SCO have some documentation on supported hardware for Unixware, and it's a pretty small list.)

khttpd (kernel web server)
irda (infrared comms)
acpi (advanced configuration and power interface)
ap1000 supercomputer
atm networking
parallel-port IDE
bluetooth
agp graphics
joysticks
Ham radio support
Rio MP3 players
i2c bus
i2o IO
firewire
gigabit networking
IBM zSeries devices
....

Looking through the list of changed files there is page after page that just can't possibly correspond to SCO code. Or are we supposed to believe that SCO spent millions of dollars writing code that was never released but still somehow leaked into Linux? I don't buy it.

Although the overall patch is pretty large, the bulk of it is changes to drivers and implementations for particular architectures. Typically drivers are maintained by people concentrating on just one type of device, fairly loosely coupled to the rest of the team.

This has two consequences for SCO, both of them bad. Firstly, if there were infringements in some drivers, those parts could be removed and replaced. People without the particular device are unaffected. It's not core code.

Secondly, and more interestingly, the change from 2.2 to 2.4 is the combined work of hundreds or thousands of programmers. There is no single programmer who contributed a million lines of code. I doubt if even the top ten most prolific contributors put in a million lines between them. The key developers are going to be reviewing other people's changes and working on core features, but not necessarily generating enormous volumes of code.

For a million lines of SCO code to get in, we would need to have a conspiracy of dozens of programmers, pretending to work independently, but actually organized and given SCO code by IBM.

It can't be just one person. If they suddenly rocked up with a million lines under their arm people might get suspicious. :-) I don't think it could even be all of the IBM programmers on the list. Without checking, I am fairly sure their total contributions are less than a million lines. It really would take a global conspiracy.

McBride has already advanced the hypothesis that everyone who objects to his campaign against the community is secretly being paid off by IBM. To substantiate the million-lines claim, they have to show a similar conspiracy dating back several years.

posted Thu 4 Sep 2003 in /issues/sco-vs-linux | link

Anything you say...

I'm a bit of a fan of The Bill, the British cop show/soap opera. One of the standard lines is the recitation of the PACE warning, kind of like Miranda:

Anything you say will be taken down as evidence and may be used against you in a court of law

One of the awesome things about the way bloggers are defending Linux is that every unfounded and damaging statement SCO makes is being permanently recorded and carefully investigated, and laid out like scalpels on a steel tray for any lawyer who happens to take SCO on, whether they are from IBM, Red Hat, the SEC, the Australian Consumer and Competition Commission, ...

Keep talking, Darl.

posted Thu 4 Sep 2003 in /issues/sco-vs-linux | link

Press conference results out

Fleur Doidge covers the conference:

Kieran O'Shaughnessy, managing director of SCO Group in Australia, said he could not answer any 'technical' or 'legalistic' questions about the examples of code.
'But Linux is an unauthorised derivative of Unix and there is significant Unix code in Linux ... some 1,000,000 lines,' he said.

Here we are again with this one-million ~~dollars~~ lines number.

A few hundred lines here and there might have happened through individuals bending the fair-use rules. It happens, in open source and proprietary code too. And when it happens, it's almost always quietly fixed with a small patch and perhaps an apology. When this first came out, I thought: well, maybe somebody borrowed a driver for some undocumented bit of hardware from SCO. Naughty, but not the end of the world.

But a million lines is a different matter: yes, it would be a serious infringement, but it's hard to hide. Imagine stealing a car... and trying to hide it from the police in your bedroom.

The 2.4.21 kernel comprises about 4.4 million physical lines. SCO have said that 2.2 doesn't infringe, so strictly we just need to look at the changes from 2.2 to 2.4, but let's be conservative and consider the whole lot. That means that SCO are claiming nearly one quarter of the code is copyrighted by SCO. If I pick ten files at random then one or two of them are likely to be SCO's code.

In fact we can narrow this down: Linux supports many device drivers that aren't on SCO, and also many architectures other than x86. We can also eliminate filesystems and network protocols not supported by SCO, which are many. Then there are sections of Linux which are just accidentally different to SCO, such as the system call format, or the threading model. Is it really likely that somebody would lift SCO's code and then go to all the trouble of breaking compatibility? And if they did rewrite it completely, is there really any copyright infringement?

Cutting out the bits of code that don't exist in SCO, while still giving them the benefit of the doubt, I get rather less than a million lines of code. Perhaps 500,000, depending on how you cut it.

I just don't think there *are* a million lines of common functionality between Linux and SCO. If I was starting from scratch to write something like Linux, and I had carte blanch to copy from SCO then I don't think there are a million lines I'd be able to use.

You could do this more rigorously by going through SCO's feature list and picking out the particular files in Linux that match: a driver for this IDE chip and that SCSI card and so on.

And here's another question: is SCO's code really worth 3e9/1e6 = $3000 per line? Even when to date SCO are counting physical lines including whitespace and comments, and talking about code that's roughly 20 years old and would likely not even compile on a modern machine? If I had USD $3000 for every blank line I wrote...

posted Thu 4 Sep 2003 in /issues/sco-vs-linux | link

SCO Panel discussion at AUUG

Groggy reports from the AUUG Conference:

Kieran O'Shaughnessy of SCO Australia came along for a panel session with Con Zymaris and myself. Not surprisingly, the session was well attended. Kieran brought out the party line with some ridiculous number of claimed lines of infringement in the Linux kernel, so I brought out a display of the BPF code of their second example presented on 18 August, then paged back to the beginning to show the BSD license there. The look on Kieran's face really made my day. He claimed, rather stupidly, that these were just examples of the techniques that they used to find similar code. Greg Rose joined in the fun and pointed out that some of his code, written in 1976 (still at least three years after the first example) had been included in UNIX without his consent, and asked whether Kieran thought he should approach IBM for royalties. Finally I got to do the summing up, where I determined that SCO had done nothing to show any abuse of its rights, but that the code in question had indicated that their code was way out of date, and that it looked likely that they had abused the BSD license (applause).

BSD too, eh? I wish I'd been there.

If this is an example of their technique then their claims of thousands of other infringements seem pretty questionable.

posted Thu 4 Sep 2003 in /issues/sco-vs-linux | link

No refunds if SCO are lying

From vnunet:

McBride also poured cold water on the notion that Linux customers would receive refunds for licences bought should SCO lose its pending case to prove that its intellectual property (IP) is in Linux.
"We have not built in any refunds in our licensing model. The product is there and it's being used," he said.

I wonder how this might play into an ACCC complaint? Selling a licence that people probably don't need is pretty bad; ruling out refunds if it turns out to have been falsely advertized is worse.

posted Thu 4 Sep 2003 in /issues/sco-vs-linux | link

aj on Hanson's Sentence

For someone I mostly agree with, aj is good at goading me into writing. Anthony writes, on the issue of Ettridge and Hanson's sentencing:

Maybe it's just me, but I don't see a lot of point getting worked up when a thief pays back everything they stole. If you're going to slap someone on the wrists, do it to the people who can and do remedy the harm they cause.

I think it's just you.

Giving back ill-gotten gains when somebody is caught or about to be caught does not undo the crime. There are at least three reasons: Hanson had the benefit of the money in the interim, the fraud was wrong in itself regardless of whether it was profitable, and there must be a disincentive to people “trying it on”. It's right and normal for criminals to do more than just repay the money. In passing sentence Wolfe CJ did consider their attempt to repay, among other mitigating circumstances, which is also entirely reasonable.

Most people are focussing on the money, but I think it's not the main problem. $500,000 out of a national budget is significant but not all that much. However, illegally interfering with an election deserves a ton of bricks, if not two.

One Nation's fraud brought them non-monetary benefits which they can't repay. We can't go back now and re-run the election with One Nation not registered. The sentence says:

There is no doubt that the registration of the party gave it an advantage. Before the 1998 election a coalition government held government with the help of one independent. After the election, when eleven One Nation candidates were elected, a Labour government held government with the help of one independent. I cannot speculate what might have happened had each One Nation candidate gone to the election without the benefit of registration. They would have been able and permitted to advertise as One Nation candidates but they would have had to lodge their candidacies for election independently and they would have been entitled to claim electoral funding if they had qualified with the four per cent. They would not have had their names identified with One Nation, the political party on the ballot paper.

Another argument offered in their defense is that it was a mistake on a technicality and there was no criminal intent. Similar, perhaps, to accidentally filing an incorrect tax return, rather than deliberately evading tax. But a jury in an apparently fair trial found that Hanson and Ettridge knew, and should have known, that what they were doing was wrong. So that defense is out.

Mediawatch quotes the Courier Mail saying:

Hanson's jailing for not properly understanding the detail will only confirm in the minds of many of her supporters that the world is indeed stacked against them.

What it ought to confirm is that before dealing with large amounts of money or running for government you damn well should understand the detail, or consult someone who does.

Arguments about whether One Nation is unaccountable to its members or has wacky policies are beside the point. If the Democrats or the Greens managed to defraud the public purse and get false information put onto ballots, thereby potentially changing the outcome of a state election, then they ought to be hammered too.

Are people complaining about this seriously suggesting that deliberately attempting to steal votes ought to get only a slap on the wrist?

Now if it turned out that Hanson didn't get a fair trial that would be different...

posted Thu 4 Sep 2003 in /issues/politics | link

Debian Packages for OpenPegasus

I took a stab at debs for the OpenPegasus WBEM/CIM package. Not too hard, though my packages are still pretty rough.

posted Wed 3 Sep 2003 in /projects/openpegasus | link

Overblown?

A reader points out that Lucky may be seeing more malice than is really present.

posted Wed 3 Sep 2003 in /issues/tcpa | link

Summary of Principles for User-Interface Design

Summary of Principles for User-Interface Design by Talin. Many other people have written similar things, but this is a pretty good brief summary.

posted Wed 3 Sep 2003 in /software/ui | link

students.org.au

What do you suppose a domain like students.org.au might contain? Some complaints about HECS and higher education funding? Actually far more entertaining...

You might think that you are too busy to have children. You might be in the process of becoming a tycoon like Bill Gates. Well, it only takes ten minutes to get a woman pregnant, and surely you can spare ten minutes. Contrary to the "politically correct" view, you don't need to spend "quality time" with your kids. Your wife can do that, or you can hire servants. [...]
The reason that the medical profession is labelling people as mentally ill is the same as the reason they are telling kids that masturbation is normal. Apparently they are trying to create a class of people who are unemployable due to being considered mentally ill or wankers. At present, the Australian population consists of two-thirds mainstream and one-third Catholics. After they have diagnosed 20% of the population as mentally ill, the Australian population will consist of one-third mainstream, one-third unemployable, and one-third Catholics. The Catholics will no longer be a minority group.[...]
The way around this problem is to be very careful about who you have as a doctor. Don't have a doctor who is Catholic or Jewish. When you go to see a doctor for the first time, ask him what schools he went to. If he is mainstream, he will tell you, and you should check to make sure that none of them is a Catholic school. If the doctor objects to telling you this, it is a sign that he is a Catholic, as they believe that their personal history is none of your business. Another clue that a doctor is a Catholic is if he, or one of his colleagues, has an Irish, Italian or Spanish name. Never talk to doctors about personal or emotional problems, as that is outside their proper field of expertise.[...]
The present corrupt legal system will continue until an Australian government amends the High Court of Australia Act to increase the number of High Court judges. Then the government could appoint a number of honest judges, who would outnumber the existing corrupt judges. The corrupt judges could then be charged with treason, over their involvement in the corrupt Mabo case, where the judges stole government land and gave it to the Aborigines. Then the judges could be put to death for treason, and we would be back to the British system of justice.

posted Tue 2 Sep 2003 in /random | link

Lucky Green on TCPA

cypherpunks.to features slides from Lucky Green's talk on TCPA at Defcon:

As Stallman says, CBTPA = the Consume But Don't Try Programming Act. It's easy to believe his theory that the act was intentionally given an unpronounceable acronym to make discussion just slightly harder, by contrast to PATRIOT or SAFE.

posted Tue 2 Sep 2003 in /issues/tcpa | link

LinuxWorld interview with tridge

Good brief interview with Andrew Tridgell in LinuxWorld.com.au:

One of the most memorable parts of that evening was when my Linux NFS [Network File System] server died, to the point that the console seemed completely dead (the load of all those Doom WAD files obviously got to it). I was about to press reset when Linus stepped in and said he wanted to work out why it had crashed, so he could fix it. I then watched in complete amazement as Linus exploited a remote file truncation bug he knew about in the NFS server I was running which allowed him to peek into the proc filesystem on the apparently dead server and work out enough to find the bug. Up till then I had considered myself to be a pretty good programmer, and quite good at debugging system crashes, but that incident taught me that I would always be an also-ran who just isn't in the same league as people like Linus.

posted Tue 2 Sep 2003 in /software/people | link

aj on arch

aj is also looking at Arch.

To clarify my earlier comments: I think some of the problems with Arch at the moment are to do with the documentation not being up to date, but some are real shortcomings in the program. For example, tla what-changed --diffs leaves temporary file droppings in the current directory. Being able to get the changeset in structured form is nice, but leaving mess behind when I just look at the diffs is not. There is no built-in command to find out what changed between two revisions, even though this is probably one of the most-used commands for a version control system. (It's fine that this is not a fundamental operation and it's very clean that it's built on top of diffing trees, but there ought to be an easy interface for common operations.) I'm sure they'll fix it (or explain why I'm mistaken) in the fullness of time.

Subversion, meanwhile, is still answering the question “Is Subversion stable enough for me to use for my own projects?” with a less than reassuring “We think so!”.

I think that is more to do with modesty than it is with Subversion really being unstable. The Subversion developers on the whole seem less pugnacious than some other developers. I don't think that's a bad thing.

posted Tue 2 Sep 2003 in /software/vc/arch | link

Hanson Song revisited

On Fri, 22 Aug 2003, Mr Rudd wrote:

I don't like it, when you turn my cell about.
I don't like it, coz you let David Oldfield out.
My cellmate has been murdered,
my cellmate has been murdered,
my jail warden murdered,
my privileges just gone.
1-2-3-4-5-6-7-8, convict rubbish, convict hate,
I don't like it, when prison stripes are white,
I don't like it, when the guards spotlights are bright.
Please explain,
why can't my cellmate be coloured white?
I should talk to some prison doctors,
a coloured cellmate, it's just not right.
I don't like anything, I can't do anything about it.
But I like Bee Smith,
But I'm not gay
'cos I left my heart in old Long Bay.
I don't like anything,
except I like Chopper Reid.
Prison dance, prison dance, prison nation?
Not a chance!
Prison dance, Prison dance,
out of my bunk, out of my cell.
I don't like a screw without strings,
vinegar tits is in the wings,
and shes saying convict things
I won't cop that, no way.
I can't do anything about it.
But I like dancing,
and I like the Jazz,
'cos I left my heart in Alcatraz..

No, the whole thing is wrong and it stinks,
and I don't like it.

posted Tue 2 Sep 2003 in /random/humour | link

Best search requests

I'm happy to hear of all the people coming to my site looking for SCO software or Ken Park downloads. I should go through and grep out some more of them.

How nice that RFC2068 should have frozen the spelling Referer (sic). Remember kids, use a spellchecker before building a multi-billion-dollar information system!

Disturbing Search Requests and more along those lines from lawnpig. Superb. Don't we all just want to see sexy glirs?

I've heard of the older iBooks referred to as 'tangerine' and 'blueberry'. I've never heard of the newer white ones referred to as 'semen' colored. I hope he was surfing my site with both hands.

posted Mon 1 Sep 2003 in /blogs | link

DMTF links

Random management stuff:

WBEM = Web-based enterprise management. Something like SNMP, in that it allows you to find out the status of devices and software and set their configuration. Done over HTTP/SSL and XML, so possibly more secure and easier to implement.

The model has three layers: providers know how to talk to managed hardware or software. A broker or object manager (CIMOM) collates all this information, and offers it to clients.

CIM = Common Information Model. A kind of metadata structure defining the types of entitities you can talk about, with scope for extensions.

WMI and WBEM from Microsoft.

WBEM/CMI introductory slideshow (PDF).

Only tangentially related is monit, a program for monitoring and managing daemons or other programs on Unix. All the WBEM stuff is far more ambitious in scope and design, and probably far harder to actually use at this time.

posted Mon 1 Sep 2003 in /software/dmtf | link

Conclusions on arch

Looks very promising; still too early to actually use unless you really like dogfood. Some of this might be just due to problems with documentation: simple operations like "what changed since patch-32" are presumably supported but I can't work out a good way to get it out.

posted Mon 1 Sep 2003 in /software/vc/arch | link

Archives 2008: Apr Feb 2007: Jul May Feb Jan 2006: Dec Nov Oct Sep Aug Jul Jun Jan 2005: Sep Aug Jul Jun May Apr Mar Feb Jan 2004: Dec Nov Oct Sep Aug Jul Jun May Apr Mar Feb Jan 2003: Dec Nov Oct Sep Aug Jul Jun May