on the edge

computers & technology, books & writing, civilisation & society, cars & stuff

Greg Black

gjb at gbch dot net
Home page
Blog front page

If you’re not living life on the edge, you’re taking up too much space.

FQE30 at speed

Syndication / Categories


Worthy organisations

Amnesty International Australia — global defenders of human rights

global defenders of human rights

Médecins Sans Frontières — help us save lives around the world

Médecins Sans Frontières - help us save lives around the world

Electronic Frontiers Australia — protecting and promoting on-line civil liberties in Australia

Electronic Frontiers Australia


(Coming soon…)

Software resources

GNU Emacs


The FreeBSD Project

Thu, 31 Aug 2006

Another Apple recall

Shortly after I got my Powerbook, just over two years ago, Apple recalled its battery because of some fault that I no longer recall. This week, I discovered that the replacement battery is now subject to a recall. So I have to order my replacement, wait for six weeks for it to become available, respond in person within 48 hours of being informed that my new battery is ready and take care in the meantime not to leave the machine on with its battery installed unless it is under supervision. That’s a pain. It’s a bigger pain that I’m looking at a recall for the same component that has been recalled once already. I just wanted to whine on the record.

Tue, 29 Aug 2006

Lying to customers is so easy

A couple of years ago, I wrote about a customer of mine who had problems with their ADSL provisioning. For much of this month, the same customer has had further problems with the same so-called service.

The trouble started around 11 August when their ISP performed some unspecified maintenance on the ADSL services in the region where my customer has their offices. In the middle of the night, their connection went down. They complained to the ISP, who promptly blamed Telstra. Neither the ISP nor Telstra wanted to do anything about the problem.

After some days, and several followup calls, the ISP stated that the problem was definitely in the customer’s ADSL modem. The customer shipped the modem to me. I plugged it in as a replacement for a different breed of ADSL modem attached to a computer running the same OS and ancillary software to connect to the same ISP and it just worked. Clearly, the problem was not the modem. To be on the safe side, I then performed a hardware reset on the modem, reconfigured it as required, and returned it to the customer. It did not solve the problem.

The customer then tried to insist on some service from the ISP, but found things difficult when the ISP made it clear that they do not do house calls under any circumstances. Eventually, the ISP told the customer that they would check the modem if the customer attached it to a Windows computer and followed instructions. The customer sourced a Windows computer, plugged things in as required, followed instructions (including another hardware reset of the modem), and then the ISP said, “There, that’s all it needed; we just had to fix the modem and now it’s fine.”

It’s clear enough to me that they did nothing to the modem (beyond getting the customer to break its configuration for our purposes) and instead finally got around to fixing whatever they had fucked up at their end. Of course, this meant that the modem still didn’t work when returned to its normal position and it was returned to me to be reconfigured a second time. When it finally arrived back at the customer premises, it was plugged in, powered up and just worked.

By then, it was the 24th. So they were off the net for two weeks. There was no apology from the ISP. There was no offer of any reduction in the exorbitant monthly service fees they charged. The ISP lied to their customer and stuck to their lies. I’m not going to name either the ISP or the customer, as the argument is between them and it’s none of my business—other than as an astonished bystander. But I will certainly recommend against that ISP if people ever ask me for my opinions about possible service providers.

Wed, 17 May 2006

AMD experiment

I thought I should mark this blog’s arrival at the ripe old age of two years by writing something. The last couple of weeks have been filled with computer hardware dramas—ranging from the ongoing flakiness of my Powerbook (now revealed as largely the result of a dead disk) to the sudden catatonia of my main workstation.

I’ve been thinking about moving to a 64-bit platform for a while, not so much because I need the performance but because it seems like a good idea to expose my own software to any possible bugs that might be revealed by the change. And since FreeBSD, my preferred operating system, now treats the AMD 64-bit platform as a tier 1 (i.e., fully supported) platform, it seems like a good chance to try out some new gear in the form of an AMD Athlon 64 X2 Dual Core 3800+ CPU.

Somewhat to my surprise, after a simple and very fast system install, I ran into trouble with ported applications—normally one of the strengths of FreeBSD. For a large number of things I wanted to install, my attempts were met by messages chiding me for trying to build something that was i386-only on an AMD platform. This list includes big things like OpenOffice (apparently because it depends on Java, which is also i386-only) and small things like the Darcs revision control system. That was all pretty irritating.

There are claims that one can run i386 binaries on the FreeBSD AMD64 platform, but my early attempts to do that with pretty trivial programs have not met with any success so far. So I don’t hold out much hope for monsters like OpenOffice.

Since the software issues looked like being slow to resolve, I thought I should at least get as much as possible of what I had already installed going. And that led to an attempt to start X11. What a mess that turned out to be.

I stupidly thought I could use the Xorg server’s -configure option to create an initial xorg.conf file. It created the file OK. But any attempt to run it with that config file resulted in a machine that was so badly locked up that it even ignored the reset button and required a power cycle before it returned to service. And this all happened so fast that it didn’t write anything in any log files, so there was no clue about a solution.

Having proved that this behaviour was completely repeatable, I then tried the old text-mode xorgconfig program. I gave the obvious answers to all the questions, created the xorg.conf file, started X and all was fine. Why this was the only way to get there is beyond me.

In the next few days, I expect to get most of the software I use all the time running on the Athlon box and to start using it for my work. I’d be surprised if that all turns out to be plain sailing, and expect to revisit this topic over the next few weeks.

Sat, 05 Nov 2005

More on the Sony DRM story

I’m not going to keep coming back to this story, as it’s now starting to get widespread coverage—at least in the blogosphere. However, Barry Ritholtz has an interesting take on it in his blog and it seemed worth noting here.

Fri, 04 Nov 2005

Sony and the music industry go too far

Mark Russinovich’s blog entry about Sony’s disgraceful DRM misfeatures on new CDs has had wide currency in the blogosphere—which is my justification for not providing a zillion links here—but little coverage in the mainstream media as yet.

I don’t have much to add to the torrent of criticism, beyond mentioning that I’m planning to stop buying CDs altogether, at least until the industry backs away from this scheme and declares that it won’t go down that road again. I know this won’t be a popular response, as people will still want their music. But I think it’s the only one that is consistent with the way I feel about Sony and their pals.

Wed, 29 Jun 2005

Contracting fees

There was a recent discussion—which I missed—on the Humbug IRC channel about the difficulties of choosing suitable rates to charge for contracting work. Early in my time as a contractor, I went to see my accountant to review the year’s business in preparation for a tax return and he quizzed me about my rates. I explained the basis I had used to set them, and he told me that I was flat out wrong. As far as he was concerned, the rock bottom amount that I could afford to charge—back in 1984—was $75/hour. And that was only for small customers. He said I should set a base rate for customers with 500 or more staff at $110 and for big companies (i.e., with 1000 or more staff) it should be at least $150.

When I protested that my existing customers would spit the dummy if I charged that much, he said, “you don’t need them as much as they need you and you’re no use to them if you go broke.” When I made it clear that I could not just double what I was charging, he told me to leave the existing customers on the same hourly rate, but to charge them for twice the number of hours and to put those free hours to work looking for new customers.

Despite my fears, most of my customers accepted the new rates, once I explained where the idea came from. So I’ve continued like that ever since, with obvious adjustments over the years to take account of inflation. I should point out that the starting figures above were for the case of direct contracting—if I had ever used an agency, they would have had to charge me out at considerably more.

A year or so later, I had another question for the accountant—I was starting to get lots of requests for fixed-price quotes and I was finding it hard to find a way of being fair to both sides. The accountant was clear about that too. He told me to argue strenuously against fixed-price quotes and to demand payment in advance to prepare such quotes. He suggested a flat fee of $2,000 in advance for a quote and he also suggested that I prepare the quote by carefully estimating everything and then, for simple jobs where there was no chance of getting it very wrong, to double the quote; and for non-trivial projects to triple it (or even quadruple it if in any doubt). Again, this has turned out to be very sound advice.

And just in case this has tempted anybody into asking me to do some work, I should point out that these days I also load my rates with a factor intended to discourage people—I’m certainly not looking for work at this stage of my life.

Tue, 17 May 2005

Tiger dot-1 seems like a step forward

I’ve moaned about my unhappiness with Apple’s vaunted Tiger update to OS X in several fora—in particular a remarkable propensity for all kinds of applications and the OS itself to crash and a completely broken combination of DHCP over WiFi (in place of a very reliable setup under Panther).

Today, I was alerted to the availability of un upgrade to version 10.4.1 and immediately upgraded my Powerbook with the 37+ MB bundle that promised—amongst other things—improvements to DHCP with WiFi. Of course, it’s way too early to announce that everything is perfect, but I’m pleased to be able to say that, so far, nothing has crashed and the wireless networking stuff seems to be back to its old self. This is pleasing, after all the grief I’ve had with Tiger.

It’s good to see that Apple got some fixes sorted out and available fairly quickly, although it rather astonishes me that they chose to deliver something quite so broken first. My laptop has been little more use than a paperweight for the last two weeks. So I’m willing to persevere with my Mac for a bit longer, but I’m still hostile towards Apple.

Fri, 06 May 2005

Apple support needs a wakeup call

Recently, I decided to have my Powerbook’s dodgy keyboard replaced while the warranty was still valid. At the same time, I decided to increase its memory from the clearly-inadequate 256 MB that it came with. Apple’s prices for GB chips are still outrageous, so I settled for an extra 512 MB on the grounds that 768 MB should be more than enough for a bit of web surfing, email and the odd ssh connection to machines of interest.

As it happens, the performance of the machine with the extra memory is still pretty woeful—although some of that might be a consequence of my too-hasty move from Panther to Tiger (of which more later).

What I planned to write about in the context of the hardware work was the difficulty of finding approved Apple places that were willing to be flexible and helpful in their approach to the job. I can list several places in Brisbane that I will certainly never take my business to, but I won’t as I don’t want to give them any publicity. However, having found a shop that seemed willing to help, I had planned to sing their praises once the job was complete.

Sadly, I won’t be doing that. They assured me that the parts would be in within two to three days and they told me that I had to swear on my mother’s grave to present the machine for the work to be done within 24 hours of being called. That suited me, so I happily waited for the call. After five working days, I called them. My contact was busy, but the person I spoke to promised me that he’d call me shortly. That was a Friday. I heard nothing. This was annoying, as I’d been confident that I would have the repaired machine for the coming (long) weekend.

On the Tuesday, I rang again. “Oh yes, the parts arrived on Friday.” So I offered to drive in immediately. That didn’t suit them and they put me off to the Wednesday. I arrived at the appointed time, handed over the machine and waited for the work to be done. When it was finished, I paid for the memory and went home. On arrival, I discovered that the protective sheet that had been between keyboard and screen was missing. I rang and they said I could come in and pick it up. I suggested that they could post it to me instead. They agreed and verified my postal address. It was to be posted that afternoon. It did not arrive over the next week, so I emailed them to remind them. I got no answer and it still has not arrived.

Yes, this is a small item and perhaps of no great importance, but it shows a contempt for the customer that sits badly with me. So, although they were helpful in some ways that their competitors did not even try to approach, they dropped the ball quite badly in other ways. So they don’t get any free publicity either.

As for Tiger, I’m so unimpressed with it and just want to say so here as a minor counter to all the Apple fanbois who seem to think it’s wonderful. My machine is considerably less useful than it was before—with applications crashing that did not crash under Panther and even the shutdown process hanging to the extent that a power cycle was the only way to get it to complete. Its wireless connections drop out constantly and get into weird wedged states at the drop of a hat. The silly dashboard stuff can’t be readily customised to work in Brisbane. And I could go on, but I suspect that a longer litany of complaints won’t help anybody.

Fri, 29 Apr 2005

Apple know how to annoy

When I ordered my upgrade to Mac OS X “Tiger”, I was once again annoyed to see that Apple flatly refuse to accept a PO Box for a delivery address. I can live with this to some extent—e.g., when it’s a piece of relatively expensive hardware such as laptop for which they might like to obtain a signature. But for lightweight, easily-replaced junk like an OS update, it’s plain silly.

For today’s exercise, it was more than silly; it was infuriating. They sent the parcel by Australia Post’s Express Post service which does not collect signatures and which is only too happy to deliver to PO boxes. So, rather than wonder how to manage my day so that I had a chance of being at home when the courier called, I could have just gone about my business knowing my parcel was safe at the Post Office if only Apple cared a bit about the convenience of their customers rather than the importance of rigid adherence to some stupid corporate rule. Idiots.

As for what’s in the box—which I tripped over on the front step as I headed out to do the shopping this morning—I’m not in such a hurry to find out and have not yet bothered to open it to see if the claimed contents are in fact present.

Fri, 25 Feb 2005

USB problem solved

Back in August, I wrote about problems I was having with a USB mouse. Since then, I’ve identified issues with any USB devices I’ve tried in that box. Recently, I installed FreeBSD-5.3 on another similar box and—on a whim—decided to try the “problem” mouse, and it just worked. Subsequently, I tried some other USB devices in that box and they also just worked. So I formed a tentative hypothesis that the USB support on FreeBSD-4.10 was less reliable than under 5.3.

A couple of days back, this was put to the test as I began building my new workstation, based on the motherboard discussed above. To get it running, I stole the FreeBSD boot disk from the machine where the USB stuff worked, confidently expecting success. Initially, the mouse appeared to work, but various things went wrong once I accessed the SCSI controller. This was followed by many frustrating hours over the next day and a half where I tried many combinations of hardware and BIOS settings, all to no avail.

I was also busily boring the #humbug IRC channel with my whining and eventually the suggestion was made that I should try upgrading the BIOS. This had not occurred to me at all, as the working motherboard was a slightly older version of the same thing with a slightly older BIOS—however, I finally took the plunge and upgraded the BIOS and was immediately rewarded with a working system. Why Asus had to break their BIOS on a later board is unclear to me, but the whole experience reminded me how much I hate PC hardware and everything connected with it.

At least I can now get on with my grandiose scheme of setting the machine up so that all the real things I was planning can start to get done.

Sat, 01 Jan 2005

Why is Telstra so bad at the directory business?

I frequently go to the on-line White Pages or Yellow Pages for the obvious reasons. But—for reasons which are completely incomprehensible to me—it seems to be a waste of time. Today, for instance, the White Pages could not find an entry that I had typed in exactly as it appears in the printed White Pages. And, after wasting ten minutes with the Yellow Pages looking for the same business, I eventually walked to the other end of the house and picked up the printed directories.

It has been my experience in general that it’s quicker to walk around and find the paper version, page through it and find the information I want. That’s utterly absurd. Even more absurd is the fact that, as a test, I tried looking for the same phone number with Google and had the answer in a few seconds. Telstra have nothing like the amount of data that Google have. In fact, I could provide Telstra with a faster application based on flat text files and grep. What are they doing there?

Tue, 28 Dec 2004

Apple is good and bad

Apple’s software and hardware both continue to be a matter of interest to various Humbug bloggers. I’m not going to recap the whole thing—those who have been following along already know the story and the players; others can research it if they wish.

However, I want to take up two recent fragments. First, Adrian said:

You can troop out the old stalwart anti-Apple crap all you like but it doesn’t make it true.

This, of course, is true. But it seems to miss the point. Just because somebody has a problem with Apple stuff, doesn’t make their opinions “anti-Apple crap”. Especially when we’re dealing with somebody who is technically astute and quite experienced.

Later, David says:

I fail to see a use for the hardware (and the price tag - although the laptops have come down), and I just can’t see what benefits the software offers anyone in an enterprise.

Of course, as always, I’m willing to be reeducated and ask those who can tell me why it is ready for the enterprise, or why exactly OS X is better than an appropriately setup Windows XP system to come forward and brain me.

I’m not going to comment on Windows systems as I refuse to use them. However, I certainly can’t see why anybody, either at home or in a corporate setting, would choose Apple hardware for general computing equipment, such as desktops and the like. The hardware is certainly costly and is probably no better than anything else in the commodity market (although I have no direct experience to back that up). And the general usefulness of OSX is nothing to get excited about.

So why do I have an Apple laptop? Primarily because it seemed, at the time of purchase, to be the least worst option. Ideally, I’d run FreeBSD on my laptop because that’s what I run on everything else and so it’s the simplest solution—and a solution that I am happy with. Unfortunately, my previous laptop was never quite right running either FreeBSD or Linux. I had to run FreeBSD-4.3 for my wireless network card to work because all later versions of FreeBSD had broken support for the weird chipset on that machine’s motherboard. I could run a very late Linux on it with support for the wireless card; but then suspend and general APM-related functions were completely broken. There’s more to this story, but that’s enough to give the idea.

For me, a laptop has limited functions: it needs to provide me with an ssh client, a browser, wireless networking, an IMAP client and an editor that I can use to write notes with. And it has to do dialup Internet for travel. Just about anything would do that (although the risks of Windows boxes and applications would make me reluctant to use them for these tasks). Getting free operating systems working properly on silly laptop hardware seems to be beyond the state of the art—unless you’re willing to put a lot of work into research and, even then, you risk problems as time passes. So, knowing that an Apple laptop could do what I needed and could be expected to work in harmony with the provided software, I was willing to forgo ease of use and general interoperability in the interest of making a simple purchase that could perform a simple set of tasks.

It has worked less well than I had hoped, although the fact that some of the supplied software works well with weird stuff that people send me has more or less made up for that. I’m happy to keep using it; but I expect to make various other complaints about it over the coming months. On the other hand, I’m much less happy that my sister bought one. She has found it very hard to get it to do what she wants. And I have found it very hard to provide her with support with her living remotely from me—and she gets no support from the shop where she bought the thing.

And, while I’m at it, I really hate the Windowsy way that all the Apple software works; I hate the difficulty of doing simple Unixy things; I hate the number of things that crash; I loathe the window manager (and particularly its lack of configurability); I detest the Apple knows best syndrome; and I really dislike being beholden to proprietary software.

Oh, how could I forget? I really dislike Apple fanboyism.

To conclude, Adrian also answered David’s question. I obviously disagree about the “grandmother” part of that post, but am generally in agreement with the rest. This is too long now, so I won’t belabour the point further.

Banks excel again

It wouldn’t be xmas if the banks didn’t manage to demonstrate incompetence in some fashion. Last Thursday, while visiting the local shops to collect the various food goodies that we’d ordered before going away, we attempted to pay for our purchases by credit card—as we normally do.

In three shops, the sales staff looked at us askance and mumbled stuff about difficulties with their EFTPOS machines. And, in each case, after several failed attempts, they resorted to manual transactions. However, although the manual fallback option was widespread in the past, it seems to have gone the way of the dodo—not one of those shops had a manual machine nor any stationery for manual processing. All they could do was scribble down details from the card, with prompting from us to cover the significant data that they tended to overlook in the stress of trying to do this while lots of other customers were waiting, and promise to do the right thing whenever the lines were “working” again.

Of course, the shop assistants had no knowledge of the actual problem and I would only be guessing—but it seems unlikely that this many shops would have the same problem if it was not related to some lack of capacity on the part of one or more parts of the chain involved in the provision of EFTPOS services. In the absence of any evidence at all, I’m going to blame the banks—at least until somebody sets me straight.

Wed, 17 Nov 2004

Train crash puzzle

I’ve just read a news item about the tilt train crash which contains the quote: “Queensland Rail say the train was travelling at 112 kilometres per hour along a section of track where the limit is 60kph”. This is just hearsay at present, but they do have data recorders and it’s quite possible that the quote is correct.

Although I have always been alarmed by computer systems that have the capacity to override the pilot of a passenger aircraft, on the basis that the number of factors that need to be considered is quite likely to be beyond the foresight of the programmer sitting on the ground, there’s no reason at all not to enforce speed limits on high-tech trains. After all, if you’ve assembled enough technology to get a train moving at 150 km/h, then adding such safety overrides is a trivial matter.

If the crash could have been prevented by such a system, then the responsible people include those who approved the deployment of that train without obvious safety devices.

Spam does pay

The case against recently-convicted spammer Jeremy James, as reported by The Spam Weblog, shows why spammers persist—it pays off very nicely.

Apparently James got money from between 10,000 and 17,000 individuals each month and made between $400,000 and $750,000 per month. I’m glad he’s going to jail, but I’d really like to have a meeting of minds with the idiot customers—if they’d stop responding, the spammers would stop spamming. But, for as long as there are cretins out there who will buy from spammers, the rest of us will have to put up with spam. It’s enough to make you want to go on a crusade with a baseball bat.

Firewall configuration errors

Just read an article from the June issue of Computer entitled A Quantitative Study of Firewall Configuration Errors by security researcher Avishai Wool. He has an online PDF copy available for those who aren’t members of the IEEE Computer Society.

The quantitative data are probably what make this interesting—in that they confirm what seems obvious with some useful numbers. The main conclusions are that “there are no good high-complexity rule sets” and that simplicity alone does not guarantee good results. This won’t surprise anybody, but the numbers were interesting.

Sun, 12 Sep 2004

Security confusion

In a recent post, Sandra Mansell talks about some security concerns that resulted from a breakin that she suffered:

So much for linux being more secure than Windows […] ssh is supopsed to be SECURE! Isn’t that what it stands for? Secure SHell?

The thing to recognise is that ssh is a tool whose purpose is to permit people to break into your system—so if you’re going to deploy it, then you certainly need to know how to do that safely if you don’t really want everybody in the universe to come in. If you want just one trusted user to be able to login, then configure sshd to only allow logins from that user, as an example. Consider not permitting password logins, as another suggestion. In any case, you need to study the behaviour and appropriate configuration of any software that sits on exposed ports and welcomes intruders. Stuff like apache comes to mind here, not just ssh.

Further on, Sandra mentions a few possible solutions that seem to warrant comment:

I have thought perhaps about running sshd on an unusual port to stop automated attacks against it.

Any automated attack that couldn’t find sshd running on a different port is hardly worth describing as automated. Generally, security through obscurity is not all that powerful a method.

I’ve also thought perhaps about setting up firewall rules to deny access to the ssh port to all but a few ‘trusted’ hosts.

If you know what hosts they are and you’re sure that they’ll always have the addresses you expect, that might help, a little. My experience is that the time when you need to get emergency access to a box that needs attention, then you’ll be doing it from a machine that you weren’t expecting to use, or else your remote machine will just happen to have a different address on this occasion. I’d stick to sensible restrictions based on users, not hosts.

We’re also going to be implementing an intrusion detection system. I want to know the next time somebody tries to ssh in as root.

I’d say that an IDS is probably not going to be good value in the situation described. However, I’d be using something to watch my logs and I’d want to know about any attempts to ssh in—as any user—from remote hosts.

Alternate Planet Humbug

Nice work by Brad keeping the alternate Planet Humbug online, given the woes that continue to afflict the Humbug box. Naturally, the only place where I knew to find its address was inaccessible, but I was pleased to find it with a simple Google search.

I was slightly less pleased to see that Ispell still doesn’t recognise Google. That’s true of some high profile online dictionaries as well; although Wiktionary have an entry, as does Dictionary.com, so the world will go on without Webster and Cambridge.

Wed, 08 Sep 2004

The end of an era

For many years, as a member of the ACM, the IEEE and the IEEE Computer Society, I’ve benefited from those organisations’ provision of free email aliases. When I got those addresses, they provided a benefit in the form of a permanent personal address and helped to insulate me from possible loss of other addresses; but I’ve had my own domains for many years now and there’s no real reason to fear their loss, so that’s less of a benefit. And, with the rising tide of spam, the aliases are a problem. I can’t block email addressed to them without breaking the entire system because a refusal to accept the offending email by my server would then see the forwarding system attempting to send a bounce message to the spammers and that leads to madness.

So, with some reluctance, I’ve been weaning my contacts off those aliases for the past couple of years and have now asked the organisations to cancel them. I believe everybody now knows that the aliases have been cancelled, but this is to confirm it for anybody who has missed out on hearing. As of now, the aliases gjb@acm.org, gjb@ieee.org and gjb@computer.org are no longer a means to contact me. I suppose this also means I will have to generate a new PGP key without mention of those identities.

Although I was able to cancel two of them easily on a web form, the third one required several emails and even when they cancelled it, the change did not take effect until several days later because there is a delay of 1 business day before changes to their database propagate to their servers—and this is an organisation of computer professionals.

Tue, 31 Aug 2004

Apple knows best

When I went to setup the clock on my powerbook, I was pleased to see there was an option to have it run NTP—although they don’t mention the name of the mechanism anywhere. I was less happy to see that they offered a drop down menu with just some Apple time servers in far-flung corners of the world. But then I tried to enter the NTP server on the local network by hand and it was accepted by the GUI utility and I was pleased. That lasted until I ran ntptrace, which showed clearly that my selection had just been ignored …

USB mouse dramas

I don’t know if the problem is flaky USB hardware on my relatively new motherboard, flaky USB support in FreeBSD-4.10, or something else entirely, but I have been having some real dramas with the mouse on my workstation lately. Since I get the same symptoms with various mice that continue to work as expected on other machines, I’m inclined to think the problem is not the actual mouse hardware.

Since it takes some days for the problem to manifest and since it has coincided once with a power outage, it’s been difficult to nail down all the details. However, it seems to be the case that, at some random time after a reboot, the little red light in the (optical) mouse goes out. From then on, the pointer does not respond to mouse movements. I think bad things happen if you click the mouse, but did not verify that today. On previous occasions, after discovering the mouse was playing up, I’ve tried various things with both the mouse and the keyboard and ended up wedging the machine so that—although it responded to pings—it refused to respond to anything else and had to be rebooted with a power cycle and consequent grief.

This morning, once I noted that it had failed again, I used ssh from another box to login. Then I changed the system default to look for a PS/2 mouse and halted it normally; plugged in a PS/2 adapter; and rebooted. It now remains to see if the problem recurs, but it will be days before I can be reasonably confident about that. If anybody happens to have any experience that might shed further light on this, I’d be happy to hear from you. In the meantime, I’ll just keep my fingers crossed.

Fri, 27 Aug 2004

A better shopping experience

Since I’m quick to find fault with businesses that don’t measure up to my expectations, it gives me some pleasure to report a business that did well. Many people responded to my request for suggestions about the brands of mouse they preferred—and I’m most grateful for that. The overwhelming majority suggested Microsoft; a decent minority were Logitech fans; the rest were nowhere. Having had two logitech optical mice fail recently, I decided to go with Microsoft. And since one of the recommendations—whose author shall remain nameless for now as it was in private email—included the URL for online shopping, I followed it to Harris Technology.

I have to say that I found their online shopping site hard to navigate, but I managed it. I ordered two mice with one day shipping and specified that I wanted them shipped the following day and that I wanted to be called on my mobile in advance so that I could arrange to be home to take delivery. To my surprise, they rang me half an hour later to confirm the order and delivery address details and said they would put a label on the parcel to ask the courier to ring me.

This morning, the phone rang again. The nice man from Harris explained that he’d discovered that the courier would not ring me and so he wanted to see when I’d be home. I thanked him and said I could wait until 11:00 and he assured me that the delivery would be before then. At 10:15 the courier rang to see if I was there—because there was a request on the package to do that. Minutes later, he delivered my package. And I got to sign for it on some electronic hardware, where my signature looked even worse than usual.

So, unlike Apple who promised to call and didn’t, I got three calls for this small purchase. Although I think we could have managed with fewer calls, I certainly prefer too many to too few. On the whole, I’m impressed; and I expect to take more business to Harris Technology.

The other thing that astonished me when I unpacked the first mouse was that there was a 150-page manual included. How much can there be to say about a mouse? It turns out that only 10 pages were needed to say it in English; apparently it’s cheaper to ship 150 pages to everybody than 10 pages to those who might need them. But it was the content of the last page that had me amazed. It counselled me to “be healthy”; to “eat a balanced diet and get adequate rest”; to “exercise for overall fitness”; to “learn to manage stress”; and various other motherhood suggestions. I never expected to be reading that kind of stuff in a manual that came with a mouse. Must be getting old.

Sparks and nonsense

I like to make sure my customers know I’m watching over them, so I comment on stuff that happens even when they don’t feel the need to contact me. This morning, I emailed one of them to ask why one of their boxes had been rebooted. Apparently, they’d had an electrician in doing something and he got it wrong enough that one of the four machines on a single UPS got snuffed out briefly. In the words of the customer:

You would think electricians would know what to turn off before working with electricity—not so!

Good to see that clumsiness and stupidity are not solely the province of the computer industry (or whatever it’s called this week).

Thu, 26 Aug 2004

Powerbook arrives

Yesterday, I had a bit of a whine about the difficulties I experienced with a purchase from Apple. At least the nice person I spoke to promised that she would call me to discuss suitable times to deliver my new toy—and that I could expect it mid to late next week. So of course it arrived this afternoon at the same time as one of my wife’s psychotherapy clients, meaning she had to take delivery while putting her client at ease. No, there was no phone call. Yes, my wife was not pleased. No, I am not surprised.

Now I have to wade through the horrible business of setting the thing up so that I can use it. When I looked at one of the original Macs for the first time, I was struck by how easy it was to do some of the obvious things and how it was almost impossible to do the things I was interested in. Not much has changed. I admit I got it a DHCP lease on the home LAN very easily, and I’ve been playing with Safari for a while. I think I’d rather use something else, but I’m not yet game to attempt that and I’m really waiting to get a proper mouse before I make too many decisions. It is pretty and I expect that I’ll eventually learn how to use it and that it will become something I’m glad I’ve bought—but it’s not as easy to use as I had hoped. I suppose the thing I like least is that it reminds me of using Windows and that’s not something I recall with any fondness.

Wed, 25 Aug 2004

Unexpected pitfalls with online purchasing

I’ve been trying to buy something online from the Apple store this week and have been surprised by how hard they make it. I’m not an online shopping junkie, but I’ve probably bought stuff this way about 50 times, so I’m not a complete newbie.

Problem number 1 is the website design—it’s hard to find the way in to the part I wanted; once there, if you get distracted by a customer on the phone, it times out; so you have to go back to the start and click your way back through the same million slow pages to get back to where you were.

When you finally get through all the process of making the order, they don’t tell you that you need to check back to see if they’ve processed it. I only checked because I’m in a real hurry for this item. That’s when I found that the order was suspended because they were “waiting for more information”. There was no clue about the nature of the additional information, nor had they made any attempt to email me to tell me. So I played in call centre hell for a while and eventually discovered that they felt entitled to some photo id. I happen to think that they have no right to that, but expediency won out over my natural inclinations, and I agreed to fax them something they deemed acceptable.

Since I only need a scanner once every three years, I don’t have one. And since I usually need to make thousands of photocopies at a time when I need copies, I don’t have a photocopier. So I took the documents to the local shopping centre, photocopied them, wrote the order number on the copy and faxed it to the Apple fax number. The transmission took longer than I’d have expected, but the fax machine reported success, so I was happy.

From time to time over the next few hours, I checked the website to see if the order status had changed. Apart from the slow and infuriating process of getting to the page I needed, I also found it irritating that their clock is set wrong; and that it’s in a different time zone; and that they display dates in a foreign format; and that you really have to ring them up to find out anything useful.

When I ran out of patience again, I called and spoke to the same woman I’d spoken to earlier. She asked me if I’d sent the fax. I said that it had been sent three hours earlier. She put me on hold; when she came back, she said, “Please hold a bit longer. I’ve put paper in the fax machine and there are lots of faxes coming through now.” What kind of business requires customers to send faxes and can’t be bothered keeping paper in the fax machine? For that matter, why does a high-tech company even use paper for incoming faxes?

Finally, my fax arrived and she told me that it was fine and that she would ship the order shortly. So all I have to do now is wait for the phone call to make sure that I’ll be at home when the delivery attempt is made—she promised me that they’d call, and naturally I believe her.

And just as I was about to publish this, I got a confirmation email from Apple telling me that the order had been shipped. So does the email comes from something sensible, like “Apple Orders”? Oh no, they must want it to look like spam—it’s from “Email Processor” and the subject is “Shipment notification for order # 1234567890”. I delete email that looks like that by the hundred; it was pure chance that I happened to have a window with a filtered mail log running on the same desktop I’m using right now and noticed “APPLE.COM” passing before my eyes …

Wed, 18 Aug 2004


I just opened one of the seeming millions of books that clutter up my study awaiting review and glanced at the list of recommended hardware while checking to see if the book was of any interest to me or those I write for.

There were two lists—one for required hardware and another for recommended hardware. Not surprisingly, both lists contained specifications for CPU, RAM, disk storage, etc. The part that surprised me, given that there were similar software lists further down the page, was that each hardware list specified a version of Microsoft Windows. Somehow, an author who thinks Windows is hardware seems unlikely to have anything to say that I’ll find interesting; that book will end up on the pile of stuff to be given away.

Thu, 22 Jul 2004

The law is an ass

The stupidity of our legal systems seems to know no bounds. There’s a story in The Register that states:

The UK High Court has judged that the sale, advertisement, possession for commercial purposes and use of PlayStation 2 modification chips is illegal […]

To balance that slightly, it seems that the relevant Italian court has gone the other way, recognising that it’s up to the owners of these things to determine how they will use them.

I’m not sure what it is that makes lawmakers think they deserve respect, either for themselves or their laws or those who enforce their laws, when the laws they create have such absurd interpretations—clearly, the task of the law is to protect the most wealthy and powerful while riding roughshod over the rest of us.

Wed, 21 Jul 2004

Newer technology is not necessarily better

That title is probably self-evident to most people most of the time, but there seems to be a widespread habit of forging ahead with the latest great idea without taking account of the costs. My recent problem with being locked out of my car got me thinking about this again today.

My old BMW also has central locking with electric-powered locks on the doors. But the key can be placed in the door lock and used to mechanically open a single door, regardless of the availability of power. That makes sense. The problem car, however, has locks that only work if there’s power. Possession of the key is useless if the battery is flat.

Surely it would have been easy enough to provide a workable solution from the following list of possibilities:

  • fit a door lock that works mechanically;
  • provide a key-operated release catch for the bonnet to allow connection of another battery in parallel;
  • provide an electrical connection in the boot (which can be opened with a key) to which a fresh battery can be connected.

There are probably as many other ideas that would have worked; the thing is, nobody in the design team bothered to provide a solution. Maybe they did think about it and decided that the cost of doing any of these things was greater than that of replacing the occasional window when they had to smash their way into a disabled car. I prefer to think they just didn’t think it through, but I have no evidence either way.

There’s a lot of other useless stuff on modern cars that makes them more vulnerable to problems and less easy to manage. Power steering is a great example. No car on earth needs it. No car that has it is drivable when the pump belt fails. Automatic transmissions are another case. They are reliable, up to a point—although I’ve had at least five experiences with cars that had an oil line to the radiator fail and disable the car. But, on the inevitable occasion when you go to start the car and the battery is not up to the task, you can’t give it a push and get going; you need help. This is silly. I could go on. But this seems like a good place to stop.

Thu, 15 Jul 2004

How hard can it be?

Yesterday, I flew to Melbourne for my mother’s funeral. Naturally, while I had my back turned, the wheels fell off things at home. Around 1600, my wife rang me to say that her email wasn’t working. Tricky to diagnose from where I was, so it had to wait until I got home after a rather long day. At 2030, I check the network and see that we have fallen off the Internet.

Rebooting the ADSL modem, always an astonishingly slow process, brought no relief. So I plugged in a known good handset and found no dial tone. Telstra said they’d get somebody to call me. While waiting for that call this morning, I wandered outside and saw three Telstra vans parked in the street outside the new house that’s being built. Three vans for one house? Wow.

When the man rang to tell me that he hoped to get a technician out by close of business on Friday, I thought I’d mention the boys already in the street. “Oh your problem has nothing to do with them,” he said. I suggested that they probably caused my problem, but to no avail. At 1300, Steve from Telstra rang to say that he’d be on the job soon. At 1400, we lost dial tone from the voice line as well. At 1410, Steve rang again to say that it was indeed the Telstra boys down the street who’d chopped us off and that he’d made a “temporary repair” which would be fixed properly next week.

The last “temporary repair” was made three years ago. That time, they said they’d re-cable the street urgently as there was no spare capacity to cope with faults. The street still has not been re-wired. Every time it rains, we have faults in the pit at the bottom of the hill when it fills with water. Every time we report those faults, they refuse to look in that pit until they’ve checked all the impossible sources of trouble first. And they have no corporate memory of any of this. As for me, all I can do is whine…

Wed, 30 Jun 2004

How to get from FreeBSD to Microsoft in one step

I was just at Amazon looking at the advance notice for the latest in the long-running series on the design and implementation of BSD, The Design and Implementation of the FreeBSD Operating System by Kirk McKusick and George Neville-Neil. On the lower part of the screen, there’s a standard Amazon heading: Customers interested in this title may also be interested in … and the very first item under that is:

  • Operating System
    Weighing up Microsoft and Linux? Learn more: Get the Facts

Of course, on the Amazon page, there’s actually a link to the Microsoft site in the Operating System header, but I see no reason to make it one here.

I always have problems actually sending money to Amazon—I have problems with their spamming me. But this comes close to an argument to stay away altogether. If I’m looking at books on a particular topic, I really don’t want to be targeted by bloody Microsoft’s bullshit propaganda.

Sat, 26 Jun 2004

Usenix snuggles up to the dark side

I’ve been a member of Usenix for ages and continue to find value in its publications and some of its other services. Being a believer in the idea that democracy only works if people participate, I am conscientious about voting in the elections for the Usenix board.

But this year, when the election was announced, I was quite irritated to discover that the sole candidate offered to us for the position of president was Michael B. Jones from Microsoft Research. We could vote for him or abstain. (In case it’s unfamiliar to you, the Usenix method is to appoint a “Nominating Committee” whose job is to propose “a slate of board members for the membership’s consideration.”)

As is my way, I whined about this at the time, but my letter was neither acknowledged nor printed in any Usenix documents that I could find. And, not surprisingly, when the election results were announced recently, Mike Jones was elected as the next president.

In past elections, when we had candidates such as Kirk McKusick—the outgoing president—I was prepared to put up with this peculiar system. But when it allows somebody whose professional life is bound up with an organisation which wants to destroy Unix and Linux and the Open Source and Free Software world, then I think it just stinks.

Since Usenix would not let me express these thoughts inside their fora, I’ll just have to do it here. Hardly anybody will read it, but it’s the best I can do for now.

Footnote: I have nothing against Mike Jones personally. I don’t know him; I don’t know anything about his work beyond what I just discovered via Google. I don’t think he should be president of Usenix. He’s written lots of papers and worked on a variety of stuff, mostly uninteresting. He was “one of the primary technical reviewers for the POSIX threads (pthreads) standard (ISO/IEC 9945-1:1996 a.k.a. P1003.1c)” which seems like another reason to dislike him. He was a joint author of a paper that got the “best student paper award” at the Second USENIX Windows NT Symposium. This was in August 1998. So he was still a student about five minutes ago; and he’s worked at Microsoft forever. It looks worse, now that I know more. I think I’ll stop my research now. Oh, and he also … No, really, I won’t go there.

And, before anybody sets me straight, I know he’s not quite as young as I painted him above, but this is the place where I get to say what I want.

Thu, 17 Jun 2004

Patents suck, patent profiteers suck double

I’ve come to believe that all software patents should be disallowed retrospectively and that no new ones should be granted to anybody, ever. All the justifications for patents—and there are some, although most seem flimsy to me—are only valid when it comes to protections for material goods that take capital resources to create. Software, apart from the minor cost of programmer time, costs nothing to create in the first place and absolutely nothing to replicate. Patents on software stifle innovation, while giving nothing back to the wider community. And, given the laughable inability of patent examiners to detect prior art, there’s no hope that patents would ever be limited to innovative discoveries, even assuming that they deserved patent protection.

This is a long-standing irritation for me and it was brought back to my attention by an IEEE Spectrum feature article that was published today. Here are some snippets:

Acacia claims to own patents that cover virtually every aspect of transferring digitally encoded media from a server to a customer. A few examples: the downloading of songs to computers and MP3 players such as Apple’s iPod, the streaming of video to a PC, the digital distribution of motion pictures to hotel rooms, even the use of a TiVo digital video recorder.

If Acacia’s patents are valid and as broad as Acacia thinks they are, thousands of companies—including titans like Time Warner, Disney, Microsoft, and Sony—and maybe even hundreds of millions of users will have to pay Acacia directly or indirectly. Cable, satellite, and Internet service providers, video-on-demand companies, music sites, the new Web radio enterprises, pornographers—almost anyone delivering digital video or audio across a network will be liable.

As far as I’m concerned, companies that innovate, patent and go out looking for infringers to persecute are evil. Companies like Acacia, which create nothing but simply acquire patents in order to get rich by chasing licensing deals and seeking punishment in the courts, should be stamped out of existence.

How hard can it be?

About two months ago, one of my customers in Melbourne moved offices. On my advice, they had ordered new phone lines and a new ADSL setup several weeks in advance. When they arrived in the new office, the ADSL service was not ready. The ISP blamed the Telco; the Telco blamed Telstra; Telstra were unconcerned.

I told the customer that they should make the various suppliers focus on the new phone lines and possibly the ADSL provisioning at the exchange as everything else was the same—they turned things off, carried them across the street and turned them on again.

At my suggestion, they begged the suppliers of their “service” to send a technician to their premises where he would be able to see instantly that the problem was not of the customer’s making. All the “service providers” refused to provide this small service, even when offered money.

Only the ISP was actually willing to speak to the customer, although this did not extend to willingness to help. They did their own tests, which I suspect were entirely imaginary, and told the customer that the fault was in their ADSL modem and that they should purchase a new one. In despair, the customer did that. The new modem’s instructions were incomprehensible to the customer, so the device was sent to me in Brisbane. I set it up and verified that it did work with the ISP in question and returned it to Melbourne. It was installed there and immediately failed in the same way as the “faulty” modem.

The ISP then made them jump through several more hoops before finally agreeing that Telstra should check the line. Telstra said they would visit, but did not. Instead, they announced—after some weeks—that they had “re-provisioned” the service and that the customer should test it. Of course, it did not work. Eventually, after some heated talks, they agreed to send a technician out.

When the technician rang to make an appointment, he wanted to make the visit on a Saturday and was most unhappy that this was deemed impossible by the customer, on the perfectly valid grounds that the building is not open on Saturdays. They made an appointment for the following Tuesday. This was not met. After further followup, the technician arrived today (Thursday). He spent a few minutes on site and said, “Yes, there’s a fault. I’ll go back to the exchange and fix it.”

Soon after, he rang to say that they should try it again. So we turned things off and on and typed commands dictated laboriously over the phone and it all worked just fine and has been working for about eight hours now. If only they had listened to us in the beginning and sent out a technician to fix their broken service. And my customer has been paying the premium “Business ADSL” rate for two months while being restricted to a 28k modem. Wow, that was a great effort.