You can’t complain about your computer in public without some jerk telling you that all of your problems will be solved by switching to a different platform. This weekend I was at a Christmas party, and of course at some point during the evening, the conversation shifted toward computers. It seems a few Mac people convinced one of the partygoers to buy a Mac after repeated insistence that OS X is the perfect platform, and that Apple hardware was so perfect that he’d never have to worry about computer problems again. As it turns out, things only got worse post-Mac.
“Everyone I knew had a Mac, they loved Macs,” he said to me over beer and hors d’oeuvres. This is not true; I know him, I do not have a Mac, and I would not recommend one to a PC user under any circumstances. Anyway, the story continued. “So I bought a Mac, and I really didn’t like it. I told the Mac people I knew that I really didn’t like my Mac, and they told me I just needed to get used to it. Well, I’m used to it and I still don’t like the interface, I still have trouble getting software to work correctly, and it locks up so much I can hardly use it.” His wife filled in some of the details, describing what amounts to the Beach Ball Of Death.
I am no evangelist, so aside from saying that I hate Macs, I won’t try to convince people not to use them, or to use Linux or BSD, or to stay away from Vista. I’ve learned over the years that if I make a heavy recommendation to switch platforms, the people who listen to me will then depend on me for all of their migration troubles. Also, I’ve found that arguing about computer operating systems is boring, pointless, and does not change anyone’s mind about anything. It’s easier to let people make their own mistakes and learn from them the hard way. Let the platform speak for itself. So I said to this guy, “It sounds like you have a software problem. I’d save all my data to a DVD or external hard drive and reinstall the operating system.”
Ah, but they’d considered that. They’d been in to the Apple store and asked “Geniuses” (or whatever Apple calls their technicians) about the problem and they said it was probably a bad hard drive. So I explained to this couple what the symptoms of a bad hard drive were, including the fact that hard drives usually fail over time and can be difficult to detect at first. But the guy replied that he’d already been on the phone with Apple support and ran a bunch of diagnostic tests that convinced the support person that it was not a bad hard drive.
“So I called up my Mac friends and asked if they could come over and help me fix this problem, and they said that I shouldn’t be having any problems because I’m using a Mac,” he said, frustrated. “Yes,” I replied, “It seems that a lot of evangelists, especially Apple evangelists, have trouble addressing reality. But if they convinced you to buy this computer, the least they could do is make an effort to help you with this.” He agreed, but they won’t help him — they won’t even accept the fact that the computer is malfunctioning.
Evangelism of any kind is a selfish, destructive thing. Evangelists work so passionately to get people to switch to a different way of thinking, behaving, or buying, but then they slip into the shadows when things get difficult. If you want to switch to a new computing environment, don’t listen to the people who rave about it. Instead, find a good support forum or mailing list and evaluate its level of knowledge and friendliness. Discover what the most common problems with this technology are, and determine whether or not you are willing and able to handle them. It’s that support community that will be there for you in your time of need, not the jerk at the office who convinced you that life would be more enjoyable if you and he shared the same technological preferences.
I’m working on some laptop mouse reviews, and I found myself expressing a degree of surprise in discovering that Bluetooth mice still exist. Not only do they still exist, but they’re growing in capability (this laptop mouse can switch between conventional 2.4Ghz and Bluetooth wireless technologies) and popularity, and prices are coming down. It would take too long to figure out exactly how long ago Bluetooth peripherals made their debut (mid-2000?), but I know I’ve been seeing Bluetooth mice and keyboards for at least a few years. I never expected it to last this long.
To me, Bluetooth means “expensive and unsupported.” Expensive because Bluetooth peripherals don’t just cost more than their standard wireless counterparts — they cost a lot more, sometimes by a factor of two. At one point both Microsoft and Logitech had wireless keyboard and mouse desktop sets that retailed for roughly twice the price of the same products with standard wireless RF technology. Outside of special-case scenarios where there are a number of wireless devices on the same frequency (such as an office with many computers), I could not figure out why anyone would opt for the Bluetooth edition of the desktop keyboard and mouse set. I say Bluetooth means “unsupported” because no desktop computers and a tiny number of laptop computers have Bluetooth transceivers built in. You can of course get a PCI card or USB dongle to add Bluetooth capabilities, but that only adds to the cost of using the Bluetooth peripherals you bought.
When Bluetooth first hit the market, I remember reading commentary from technology analysts predicting that it would completely replace all corded computer peripherals — mice, keyboards, printers, and anything else that connects to computers — within a short period of time. So revolutionary was this secure, low-power, long-range wireless technology that it was going to change the way we arranged physical computing environments. The funny thing about outlandish predictions is that they never come true. The only aspect of Bluetooth’s history that surprises me is that it hasn’t yet been replaced by something more inexpensively implemented.
Nowadays, most people are familiar with Bluetooth because of wireless headsets for cell phones. That is, after all, the market that Bluetooth was designed in and for. I’ve thought for quite a while that Bluetooth’s moonlight career in computer peripherals is coming to an end — if it ever had much of a beginning — but now I’m seeing a new generation of portable computer peripherals (mice, mostly) that are Bluetooth-capable. I think perhaps this says more about the evolution of cell phones as computers than it does about the need for laptop mice, though.
I have about 400 bookmarks between Opera and Firefox on my workstation and laptop computers. Every once in a while I sort through them to remove dead entries and figure out if I still need to keep some of them. Despite my efforts, I usually end up deleting very few bookmarks. I’ve been trying to analyze why this is; for some reason I have an emotional attachment to links, some of them on pages that I have not fully read, some of them on pages that have information I can’t use. Some of them are even on sites I don’t particularly like.
So I thought really hard about why I have so many bookmarks, and it boils down to the fact that I am using bookmarked links to solve two problems: Finding unique and useful information in a Google search, and finding things that I didn’t know I wanted. These things are difficult to do. For instance, I found out last year that I occasionally suffer from sleep apnea. The standard medical cure for this is to wear a Darth Vader mask hooked up to a water vapor air pump when you go to sleep, which you do for the rest of your life. An unfortunate side effect is that you wake up saying things like, “There will be no one to stop us this time!” and “You may dispense with the pleasantries, Commander, I’m here to get you back on schedule.”
In all seriousness, from my frame of reference, this solution does not address the problem. The problem is that I’m not breathing correctly when I sleep, but the sleep doctors tend to assume that the problem is that I was born without a breathing machine to hook up to at night. If I buy a machine, I’m cured for as long as I use it. Imagine how miserable vacations would be without it. Naturally, my reaction was to hit Google to determine if there are alternative treatments, and to see if I could read some stories from people who have dealt with this problem before.
I spent several hours reading through the same information over and over again on different sites, all of which were at the top of the Google results. Little of the information was conclusive, which is another way of saying it was useless. The best information came from two sites, which I’ve bookmarked: this article on a preliminary medical study, and several message threads on the Apnea Support forum. Both of these led me to develop my own cure for sleep apnea, and so far it is working rather well.
There are many such bookmarks in my collection — sites that I’m worried I’ll never find again, and may need to consult later. The IBM Lotus Symphony page, which is not navigable from www.ibm.com as far as I can tell, is a perfect example. So is this Google video capture of The Century of the Self. Since I’m doomed to forget the title of that documentary, I’ll never find it again unless I bookmark it.
So the list is 400 and growing, with no end in sight. Perhaps if there were less clutter on the Web, I wouldn’t need to pack so many links into a single menu in my Web browser.
Last week Open Season (a Register podcast) featured an interview with Mark Shuttleworth; it’s definitely worth checking out. Below is a brief summary of the more interesting topics discussed.
The interview started with talk about Dell’s Ubuntu laptop and its misbehaving touchpad. This turned into predictions on how long it will be before Linux desktop software is ready for a broader user base.
Next, the participants engaged in a long discussion of the weak points of the OpenOffice.org office suite. Mark describes the current state of OO.org as being “a bit of a tragedy.” He believes that the problem with it is the bungling interaction between the “corporate folks” and the community, also remarking that potential contributors are deterred from participating due to extremely long build times for OO.org — 8 to 9 hours. He says if he doesn’t see signs of a turnaround in the next few months he’ll start looking elsewhere for Ubuntu’s included office suite.
He believes that ultimately the success of OpenOffice.org in specific, and community-developed software in general, requires that it switch to a Firefox-style model in which the core software is small, streamlined, and handled by a small team, and additional functionality can be easily added by the wider community through extensions, without the need to compile a massive amount of code or have commit access.
They also talked about Oracle’s position in the Linux ecosystem, remarking at one point they were very close to having Ubuntu certified by Oracle, but then Oracle decided it wanted its own Linux offering, which turned out to be little more than a copy of Red Hat. Shuttleworth doesn’t think that strategy is going to work.
Looking forward, the Ubuntu founder said he was very enthusiastic about the future applications of virtualization technology, believing it will “touch a whole bunch of different things.” He also remarked about consumer electronics and the mobile versions of Ubuntu, and improved collaboration with Launchpad.
The participants also spent some time discussing The European Union v. Microsoft. Shuttleworth wondered what the effective and appropriate remedies would be. He speculates that forcing Microsoft to disclose formats and protocols in an effective way (not what they’ve done so far) and perhaps requiring access to intellectual property would go a long way to toward making sure that projects like Samba and OpenOffice.org can have a level playing field on which to compete.
Interestingly, one of the interviewers asked who Ubuntu’s biggest competitors are. Shuttleworth defied expectations by stating that Ubuntu really isn’t competing against other major Linux distros — really it’s a competition between the proprietary and open source ideologies.
Interestingly, Mark Shuttleworth spends the last 10 minutes of the interview asking his interviewer questions. He really comes off as a cool and humble guy. As an example of his humility he jokes that he hates meeting people in person because they always say “I thought you were taller.”
A badly researched Yahoo News piece recently characterized open source developers’ reluctance to adopt the new GNU General Public License version 3 as creating “a rift in the open source community between idealists who believe all software should be free of charge and free to use, and pragmatists who want to see open source software make further inroads into commercial use.” There are so many things wrong in that statement that I hardly know where to begin. Is it really so difficult to understand this stuff? Yes, there is a rift in the community — if there is a single, cohesive, unanimous community at all — but it’s not for the reasons listed in this Yahoo story.
I guess I should start with the difference between open source and free software. One is a software design philosophy officially defined and maintained by a licensing cabal; the other is a set of moral values imposed through software licenses defined and maintained by a social/political movement. There has been a natural division between the two since the inception of the Open Source Initiative in 1998. The OSI has little or nothing to do with “seeing open source software make further inroads into commercial use,” per the Yahoo author’s article. The very first sentence on the OSI Web site is: “Open source is a development method for software that harnesses the power of distributed peer review and transparency of process. The promise of open source is better quality, higher reliability, more flexibility, lower cost, and an end to predatory vendor lock-in.” Where does it say there that corporate adoption is a goal? Not that getting more open source software into enterprise-grade businesses wouldn’t benefit the OSI or open source software in general, but since when has this been one of the OSI’s primary goals?
The Free Software Foundation has vociferously expressed in every possible way over the past 20+ years that the “free” in “free software” has absolutely nothing to do with price. Again: Free software has absolutely nothing to do with price; it is instead a set of moral obligations to provide certain allowances to users and programmers. It takes less than 30 seconds worth of reading on the fsf.org Web site to discover this fact. You’d have to be some kind of moron to make a mistake like this in a professional publication.
Despite the Yahoo author’s mistakes, there really is a rift forming over the GPLv3. Indeed the gap between free software-ists and open source-ists is widening with the GPLv3, mainly because of the restrictions it includes. FSF supporters say that the restrictions protect freedom; open source pragmatists say that the restrictions harm the very same freedom, though the OSI has approved GPLv3 as officially “open source.” The OSI isn’t really a leader in the open source community, despite its name — all it really does is allow people to use the OSI trademark if their software license conforms to the open source requirements. The real leaders are the people approving patches, writing code, and performing code audits — they’re the ones who truly determine what open source software is and is not.
Anyway, fundamental dichotomies aside, since the June release of GPLv3 I have seen some of the free software moderates defecting to the open source camp, too. With events like Eben Moglen’s bizarre tirade against Tim O’Reilly, the FSF leadership (even though Moglen is no longer officially a leader, he still acts as a spokesperson) is appearing more like a group of religious extremists, and less like programmers intent on creating free replacements for proprietary programs. The FSF’s focus is increasingly to protest, boycott, censure, and exclude all things it does not find morally agreeable. By contrast, it is the open source community that is doing all the real work — the coding, the documentation requesting, and the reverse-engineering. The rift is forming primarily between the people who do the talking, and the people who do the actual software development. This has come to mean a rift between free software, which for all its bluster and bombast has failed miserably to produce a complete GNU operating system; and open source, which has produced many financially and/or technologically successful operating systems over virtually the same period of time.
Wall Street Journal Columnist Walt Mossberg says that Ubuntu isn’t for mainstream computer users. He may be right, but I would like to know what his basis for comparison is. Though I can see based on his photo that Walt is decades older than I am and has had more access to more operating systems over the course of his career, I have to question his experience in installing, configuring, and using desktop operating systems. The issues he lists are genuine, but not unsolvable, and aren’t materially different than the same initial configuration trouble that any user could have with any operating system on any modern computer.
Recently I have been looking into moving to Europe — just for fun, really, because realistically this is not something I could financially afford to do. I was reading up on some advice for Americans moving to Switzerland when I came across this brilliant piece of advice: “If you find yourself getting angry, accusing the Swiss of what may appear as odd behavior to you, this is a good chance to analyse exactly what behavior sets this off and skillfully try to find new ways of communicating to establish good and lasting relationships. Keep in mind that you didn’t really like everybody ‘back home’ either.”
The same applies when you are a Windows or OS X user in Linux Land. Keep in mind that your traditional operating system did not always work as intended, and if you had to install or reinstall it from scratch, you know how difficult it was to find the most current drivers, apply hours worth of patches, readjust the ridiculous default settings to your preference, and reinstall all the software you pirated, er, legally purchased. This is no different on Linux, and in fact, I would say that it’s much easier to switch to Linux than it is to switch to Windows or OS X after you’re used to Linux. Treat this not as a reason to say that Linux sucks, but as an opportunity to learn how to properly configure and control your new operating environment.
Mossberg’s conclusion does make a lot of sense, though. If Dell is supplying the operating system, it should work perfectly from the factory. After all, when Dell gives you a Windows machine, it guarantees that it works properly in its initial condition. Why should this be any different with Ubuntu Linux? I’m suspicious of this behavior; it looks to me like sabotage, like Dell is purposefully half-assing its Linux effort so that it can “prove” that Linux isn’t a viable desktop OS, and the issue can be safely put to rest. More likely, though, it’s a lack of funding and internal commitment to making Linux desktop systems succeed.
The thing is, Linux — and certainly BSD — users don’t really want Linux preinstalled. What they want is a fully Linux- and/or BSD-compatible computer that they can install their preferred OS on. They want the equivalent of a sticker on the machine that says “FreeBSD 6.1 certified” or “Designed for Ubuntu 7.10.” We already have Windows logos that say things like this. Why can’t Dell give its customers the same benefit of knowing which of its systems works perfectly with FreeBSD and Ubuntu Linux? Based on that information, the rest of the BSD and Linux users can figure out if their preferred OS will work with it.
From Kyle Bennett over at HardOCP, I’ve learned today that I’m not the only one who has problems with some tech PR firms. This time it’s a PR guy who made ridiculous claims about a computer’s performance, then refused to back them up with facts or figures. Before you continue reading, I should mention that most PR people are professional, pleasant to deal with, and admit their limitations when it comes to technical subjects. Updated
The PR guy Bennett dealt with was David Tractenberg, who is listed as the president of Traction PR. Among Traction’s list of gaming industry clients, I see only two names I recognize — New Line and Konami. New Line being the film production company (and as we all know, movie-based games are always horrible — no exceptions), and Konami being most recognized for its Nintendo smash hit from an era before a significant portion of today’s gamers were born, Contra (“Congratulations. You have defeated the vile Red Falcon and saved the universe. Consider yourself a hero.”). Not really what I’d call “heavy hitters” in the computer and console game worlds.
More interestingly, Traction PR lists its “alumni clients” on the same page, and those are some pretty significant names — Sony and HP, for instance. But I can’t help but wonder why these are ex-clients. Perhaps it was David Tractenberg’s special brand of charm in saying such things to journalists as (according to the HardOCP article authored by Kyle Bennett), “Really, were you being serious? Just need to know if want a real answer on this one…” in response to a question asking what makes the company he’s representing a market leader. Or maybe it was the creepy phone message Tractenberg reportedly left Bennett. I’ve had one or two of those in my career. If I were on Tractenberg’s client list after seeing how he behaves in email and phone conversations with the very people he is trying to reach, I would find the quickest way to become an “alumni client” myself.
I have to wonder if some of the smaller PR agencies are just press release factories that megaspam journalists in the hope that a few of them will just repost the release without question, cringing when any degree of scrutiny is applied to their marketingspeak. By contrast, some smaller operations like Pat Meier Associates and Modena Barasch Communications are consistently top-notch in their approach — innovative, attentive, and creative in all of my dealings with them. I don’t know why, but I had this unrealistic notion that public relations is a field that is able to weed out the losers early on. I guess maybe it’s like any other industry or profession, where there are people who enjoy doing good work, and people who are just doing a job.
Addendum
I got a call from David Tractenberg this evening, asking if I could modify or remove the above post. I won’t do that unless there is a factual error or inaccuracy, but I did offer to print a response or an interview with him if he’d like. Tractenberg turned those offers down, saying that he didn’t want to make this any worse than it already was. I’m not convinced that a real news story or interview on the matter would do harm, but that’s not my decision to make.
Anyway, David’s not quite as bad as the situation suggests, for the simple reason that if he were, he wouldn’t have called. I’m sure everyone has their bad days, etc., but when you talk to the public or to someone who will report to the public, you have to be so very careful with what you say. I’m sure that there is another side to the story, and that both people are “right” from their own frame of reference. So the heart of the matter — that a situation happened and that things were said and emailed that should not have been said or emailed — remains true. As to whether this speaks badly of Traction PR, I don’t really know. Despite my reluctance to remove this post, Tractenberg still offered to send us a Commodore gaming system to review, which I will gladly do if/when a system becomes available. That puts him on a level above Apple, Mayo Communications, and the other miserable PR reps that I’ve dealt with in the past who were, to their detriment, totally unrepentant for their lapses and were not able to salvage a bad situation.
As to whether the Commodore system really does perform 30% better than some or all of its competitors, that remains to be seen, assuming I do eventually get one in to review.
For quite a while I have heard, read, and been personally told stories about messages that have been deleted from Groklaw’s comment section. Specifically these are posts that do not agree with Pamela Jones’ opinions and conclusions, or seek to debate or clarify issues that are not clear-cut or obvious. I’m curious about the voices that have been silenced by Groklaw’s censor, and might like to write a story about it if I can collect enough information about the people and posts that PJ would like us not to read. So if you have posted a message on Groklaw and subsequently found that it was removed or edited, or if your Groklaw user account has been terminated without your consent, tell me about it — email me at jem at thejemreport.com.
A few days ago I wrote about how reality trumps idealism in software licensing, the point being that there is a wide gap between what free software supporters wish the software world were like, and what the software world actually is. There is a similar disconnect between a program or machine’s actual utility and the image projected by its producers, marketers, and supporters. Nothing — but nothing — in the information technology realm is perfectly intuitive for everyone, nor does it always work correctly or lack a certain percentage of production flaws. But for fans of specific (usually underdog) technologies, there is an obsession with the image that the object of their affection “just works.”
Maybe the real issue is the blind ignorance that comes packaged with the slightly unhealthy sentiment of love. How many parents have you known who think that their astonishingly average son is the smartest kid in the world, or their plain-looking daughter is a beautiful princess? How frequently have you heard people gush about how their 6-month old baby is performing at a 10-month old level, with all the fervor of someone who wholeheartedly believes that such trivia is impressive and interesting to anyone who does not share the child’s genes? Somewhere inside, these people know that they’re hawking absurdities, and masking worries about the obvious with hope for change.
I think the same is true beyond parents and children. It’s certainly true in sports and politics, where it’s widely believed that if you yell loud enough and argue strongly enough, that the facts and probabilities will be irrevocably swayed and reality will be changed. It never is, but that hasn’t ever stopped anyone from insisting that “Our team’s gonna win!” when the odds or the facts suggest otherwise. Sadly, aficionados wrongly assume that if their sports team or political party loses, it is because they and their fellow fans did not wish hard enough.
In the technology world we have Apple fans who silently ignore the reality of myriad hardware defects and unethical business practices, Apple’s refusal to replace known defective products, and the recent incident where Apple iPhone customers got screwed out of $200, not to mention Apple’s life-long habit of abandoning platforms with no advance warning. Apple is arguably worse than Microsoft in many (if not all) ways, but something about its overpriced, unreliable fashion accessories appeals to a solid 3% of the computing market. And that 3% are cut from the same mold as the gushing parents, screaming fans, and arguing politicos.
The same applies to many (if not most) Linux and BSD fans — blind advocacy frequently overcomes the reality of the software. Too frequently that means distorting or ignoring the facts in order to puff up the object of the fans’ affection. The point of all of this blustering is partly just raw emotional expression, but a larger portion of it is evangelism. When you’re a real fan, you want to be among other fans, and you want as many of them as possible. Each conflict your team engages becomes a war between idealistically opposite factions, a struggle between good and evil, and David vs. Goliath revisited — all in one.
With computers, fans like to claim that something “just works,” meaning no manual-reading or configuration is necessary, and you’ll never have a reasonable problem with the product. That is never true, though. Everything needs maintenance, and all things degrade over time. Apple computers have all of the same problems that other computers do, and the operating system is no less bug- and problem-prone than most others. A MacBook Pro with OS X does not make you more creative than a cheap eMachines with Windows ME. Your Linux desktop computer will crash and need to be restarted sometimes. You will run into frequent permissions problems on a desktop Linux system, and you won’t always know that they’re permissions problems. Windows doesn’t work all that well with all Windows games. OpenBSD performs poorly under a lot of circumstances. FreeBSD is labor-intensive to maintain over a long period of time. There is no technology utopia; anyone who says differently is just showing you more pictures of their “beautiful” ugly kids.
A preliminary report on LWN.net indicates that AMD will soon release the hardware specifications for its graphics chips, starting with the R500 onward. If this is both true and accurate, this is the best news the open source world has had in a long time. Open source drivers can be mostly or totally useless, but open hardware specifications — those that are made available by the manufacturer and do not require a non-disclosure agreement — enable the entire open source operating system world to write good drivers.
Last year I wrote a lengthy piece about OpenBSD’s quest for hardware specifications. I link to it frequently because it shows the real front line in the open source operating system world, and explains why and how the effort to write new hardware drivers is getting much more difficult. One of the things I learned that really surprised me while doing interviews for that article and the one on the OLPC proprietary hardware fiasco (both articles came from the same set of interviews and the same body of research) was that open, complete, and accurate specifications for hardware are far more important than an open source driver for the same. You can have all of the source code to a driver, but if you don’t have the hardware specs, it is very difficult to figure out what the code does. It’s like trying to do component-level modifications to a motherboard without a schematic — you could probably do it, given experience, time, and enough resources to fail dozens of times, but why would you want to?
Secondly, open driver source code may not be any good, or directly usable by anything other than the platform it was intended for. It’s possible that the driver is horribly written, in which case you might be better off writing it from scratch anyway. While it certainly helps to be able to look at someone else’s driver code to figure out how to safely, securely, and efficiently interact with the hardware, the code for a FreeBSD driver will be almost entirely unusable in the Linux kernel, and even other BSD projects. The various abilities of each operating system kernel conform to different design philosophies, most notably in memory allocation, process scheduling, security, and threading models. That’s not to mention the free software and open source licensing issues surrounding copied driver code, which has caused much trouble lately. Aside from technical and licensing barriers, many operating system developers would simply prefer to write their own code rather than use someone else’s.
When you have open and complete hardware specs, nobody is left out. Linux is the biggest player in the open source world and is most likely to get it an open source driver before any other Unix-like operating system. Sun could probably manage to arrange something for Solaris as well, but Sun’s history says that they’d more likely pay to develop a proprietary driver instead. FreeBSD is always on the cusp — it has proprietary Nvidia drivers, for instance, but they’ve never worked as well as they do on Linux. OpenBSD doesn’t want anyone’s binary drivers, not that there are any available for it anyway, so it’s totally left out of the video driver game. Open specifications give every operating system the same, standard resources to work with — and nobody gets left out.
