Ever look at the price tag on a Textbook?, those things are expensive.
Textbooks are expensive, but in my experience the majority of books that need to tell people how many planets are in the solar system are not textbooks. People who own astronomy textbooks already know how many planets there are, and most are likely to be more interested in the fact that Pluto is a small icy double-body in the outer solar system, than in whether people classify it as a planet. If anything, the new classification will make it easier for textbook users, because it more cleanly groups the different types of bodies in the solar system... but any good textbook probably attended to that already.
Books likely to need updating are common library books, and encyclopedias (which need updating frequently for all the other information they contain, anyway). In either case, the types of people who own and use these books aren't the types of people who would care a lot if it was out of date. If this decision has any influence on book sales (textbook or otherwise) I'd be very surprised, and I'm sure the book publishers know this too.
That is, if a hardware manufacturer releases source for a software product which drives their hardware but the hardware won't actually run modified versions, that's not really "open source" (and certainly not "free software"). That's "look but don't touch".
I'm starting to agree with this too, but it's taken a while of thinking to get to this stage. The main issue I've been trying to work out, which is one that's come up in previous GPL3 discussions, is with open source voting machines, and the legality of using GPL3'd code to write software for open source voting machines that might need to be restricted to running a specific revision of specific software to be trustworthy. In that case it's really in everybody's interests that the code can't simply be changed by someone at the end of the line who controls a voting booth, and I'd like to think that secure-from-tampering hardware is still possible with GPL3.
I think the best answer to this, if it holds up legally, is that the central voting authority which owns the machines (state authority, national authority, whatever), is the authority that should be adjusting the hardware to restrict it to running a particular set of code. That said, I'm still not sure exactly how the legal side of it works out, which would probably depend on a lot of details about who owns what, who's employing who, who's giving what to who, and what counts as "distribution".
It'd be nice if Google would eventually offer an encryption system such that no plain-text documents ever saw Google's servers. (ie. It'd need to be encrypted in the client before being saved, and ideally the whole system for this would have verifiable source available.) Collaboration could be a bit of an issue, but it'd still be feasible if the people collaborating on the document could share a common key.
The biggest down side for Google in this scenario is that it wouldn't (easily) be able to scan documents for any future targeted advertising, or whatever else they might have in mind.
All this said, GMail (let alone all the other webmail out there) has already demonstrated that there are massive amounts of people out there who are happy to put personal and sensitive information on someone else's servers. (Ironically for me, I'm also one of them in the case of gmail, I guess.)
If you give the client the option between tweaking a template to a report, and tweaking the queries that feed the damn report so it runs 10% faster, the client will ask you to first make it pretty, then worry about the queries. If you dare ask them why, they will give you a b.s. explanation that it is all about perception. That the pretty page looks more "professional" and it looks like more work and care was put into it.
I know this is a nitpick -- I fully agree that clients often don't appreciate how serious some bugs are, but I don't think it'd be unreasonable for a client to want to prioritise the look of a report over making the query 10% faster. Really it'd come down to how critical the exact-ness of the template is. If it's a government organisation, for instance, there might be legislation in place that says it has to be a certain format or a law's being broken. Or it might just be an eyesore to look at. You really can't know without talking to the customer or reading the specs (if they've been well authored).
If the query was returning the wrong data, or if it took orders of magnitude longer to run than it needed to, it'd definitely be worth looking at first as a serious bug. Really though, if it's a report that a user has to sit and wait for anyway, 10% longer running time behind their other tasks probably won't have nearly as much impact on their work day as having to use a report that looks different from how they need it to look. If it's a report that runs nearly instantaneously, 10% difference is nothing anyway.
How on earth will Ubuntu be able to provide the security patch needed to immunize my kernel against a new vulnerability? It will depend on the blob provider.
I think this is an important point about open source in general, and it reaches much further than just drivers. It applies to the entire collection of applications in a distribution.
This is one of the key things that I think Open Source Software and licensing has going for it. On the surface, it allows someone to take the code and use it elsewhere, but the fundamental nature of the licenses also means that support for existing software can be transferred much more easily. This makes it a lot simpler to have a decent and reliable package management system in open source OS distributions. Essentially, it makes it possible for a single entity to support an entire system, and have the freedom to update any part of it. Even though it sometimes means minor forking of code to fit with a particular distro, it's already been shown, for a long time, that this can be done in a maintainable way. Distros can successfully tinker with code for their own distro while still keeping it in sync with upstream releases.
People often complain about linux app installation not being as simple as download, click, run, but I can't imagine how it'd be possible to have anything like yum or apt-get, let alone a stable underlying package management system like rpm or dpkg, in a Windows environment with closed source licenses. Windows has a built-in package manager, but it's very basic compared with others. Personally I've always had problems trusting that installers are doing what I hope it's doing... keeping in mind that most application installers tend to come from third parties who might or might not know what they're doing. When the installation package comes via a central provider who's certified it for use with their system, it's much more trustworthy.
It's simply not possible for Microsoft, or any other organisation, to collect together thousands of third party Windows applications and provide them in a stable, integrated repository that's supported by a single provider. The closed source licenses of software vendors typically prevent this, and they also typically prevent anyone from tinkering with source code and/or re-releasing the product, if source code is even available at all.
At best, Microsoft can provide some level of stability with Microsoft provided applications (Office, IE, some specific dev tools, etc), but after that the user is on their own, and has to get support from elsewhere. To top it off, Microsoft's own closed source policies for Windows means that third party supporters don't have full control to keep their apps stable, either. (eg. Any Windows bugs that affect an application can only be fixed at the whim of Microsoft, fixes might break other apps that a user needs, and so on.) As soon as a user wants to use something that Microsoft doesn't provide, or wants to do something in a way other than the one and only Microsoft way, it's necessary to go to a third party, and then it risks becoming a support nightmare.
I use open source software for all sorts of reasons, but I think most of them trace back to this basic thing that the licenses of individual bits of software allow for my entire system to be much more stable. I use two binary blob drivers on my laptop (nvidia and broadcom) because I have to to get it to work how I want it to, but the next laptop I buy will definitely be scrutinised much more for its hardware -- both drivers have just been such a hassle to maintain in a way that keeps the system stable.
Free Software is not about a development method but about a way of licensing software.
I agree. The key for free software (to me) is not whether the developer(s) are engaging with a community to develop the software -- it's whether people in the community can freely fork the code and continue to develop it independently if they decide they don't like the way that the original developer(s) are going.
As long as Open Solaris is licensed in such a way, it's fine by me to refer to it open source.
Sad to see that even such a big company with such a big "linux-centre" like IBM doesn't really understand Free Software.
More likely it's just one part of the company not being in-tune with another part of the company. I'm sure there are lots of people at IBM who'd understand the main points of OSS. It's a shame that this got as far as an executive's press release, though. It implies that the people at the top of IBM's open source project don't really get it, even if others do.
Have any of them actually read I, Robot? I swear to god, am I in some tiny minority who doesn't believe that this book was all about promulgating the infallible virtue of these three laws, but was instead a series of parables about the failings that result from codifying morality into inflexible dogma?
Perhaps. Whether this is true or not, I'm certain that Asimov himself is on record as saying (probably in one of his intros) that the Three Laws are primarily a story device for interesting fiction. When he got started writing robot stories, he was most interested in writing stories that weren't about robotic terror machines going crazy. Many people were afraid of and felt threatened by robots in the 1950's.
Developing the Three Laws of Robotics and treating them as un-breakable was a convenient way to put boundaries on things (without having to explain their implementation). It made it easier to justify his robots acting the way they did, creating new and interesting stories as a result. By the end of it, Asimov was also on record as saying something along the lines of that he thought he'd probably explored all the possibilities of how robots might react in different situations if they were constrained by these laws.
Asimov was probably one of the earliest people to recognise that robots and machines could be beneficial to humanity, and didn't need to replace humanity. He got his message across by convincing people that they could be built with the primary intention of helping people. Trying to apply the laws to reality, however, is asking for trouble.
Give Dvorak articles a category, please!
on
Dvorak Adores YouTube
·
· Score: 4, Informative
How long until Slashdot can get a category for articles by Dvorak? I'd really like the opportunity to be able to filter his stories from my front page.
I realise that your complaint is mostly about the lying aspect and that you probably weren't meaning to imply that propaganda is illegal. Assuming there are lies intended to demean someone without justification then by all means let it go through the courts, but I think it's worth making it clear that propaganda doesn't imply lying or illegality, paid or not. In fact, the word itself means little more than the spreading of views and interests that promote a particular cause, influence people's opinions, and whatever else.
This is NOT democracy. Anyone who tells that this is democracy, are probably other paid propagandists.
I haven't made up my mind on this issue. Part of me wants to agree that democracy should include (by definition) free access to reliable and objective information about important issues that isn't influenced by propaganda, but realistically that information still is available to anyone who chooses to look for it. On the other hand, people pushing their ideas and trying to influence opinion is a side effect of free speech -- democracy invites it as much as something like communism probably invites government official corruption. I'm not sure how it can really be changed without some dramatic changes to the way government works. Personally I'm fortunate to live in a relatively small country, and I get the impression that the result is a national government which is much more open and less like a dinosaur than the US Federal Government. Maybe it's just that democracy doesn't scale well, or that the US form of democracy hasn't scaled well.
Users don't care about release deadlines (other than some who want the latest toys ASAP). The only people who care about Vista release "deadlines" are corporate stock holders. There's no value in rushing it out if you end up angering customers who may later switch to another vendor.
We definitely care about it at my workplace, because it's likely that the next desktop we roll out to all our users will be a Vista desktop. We're not exactly going to roll it out immediately on release, but we do want a good idea of how long the final product will have been out there and tested before we use it on a large scale ourselves.
A 3 or 4 month delay in the Vista release probably means we'll have to factor in a 3 or 4 month delay in our own rollout schedule, just so we can have enough time to play with it properly (post-beta), check everything works, wait for other vendors to sort out their apps to fit Vista, and sort out problems with our specialised apps. This affects all kinds of things in the running of the business, and on short notice a delay can be a real annoyance. (One that we're prepared for given Microsoft's repution, but an annoyance all the same.)
RealNetworks will be packaging Firefox with their software NOT the other way around.
This is true, and good for them for reaching an agreement, although I'm curious to know what the hooks are in the agreement since Real would probably have been able to take and re-package Firefox regardless given its licensing. The main thing that I'm wondering about this is how much and in what ways Real might plan to modify Firefox for their own packaged version.
Will people with the Real build of Firefox get selective popup advertising from Real.Com? Will automatic upgrades come from the official Firefox distribution site, or from Real.Com? Will people easily be able to revert their installation to a generic Firefox? Given Real's reputation and past actions, it wouldn't surprise me too much if this became an issue.
This move really underscores the rift in the Open Source community as to what the goal of Open Source really is. Should we be spreading a philosophy, or just trying to get as many people using our favorite software as possible?
I don't know about you, but I don't subscribe to either of these.
I consider myself part of the Open Source community because I both use Open Source, and from time to time I've also written and released my own Open Source. I don't particularly care about spreading the philosophy (although I'm happy to explain it to people), and I don't feel the need to make people use it (although I'm happy to help them if they want to, within reason).
Personally I like and use Open Source software because in the ways that I like using software, I find it to be of superior quality and better suited to my needs for a variety of reasons. Running campaigns and trying to convert people to new philosophies has nothing to do with it.
Individual people or organisations within the open source community might have goals, but I don't think it's a serious problem if different groups disagree. I'm also not sure if it's meaningful to claim that people should be aiming for a goal just because they're involved in open source. If anything, perhaps one issue that could be addressed is how to better identify different interest groups without trying to bundle them all into the "Open Source Software Community" basket.
Then I guess the GPL v3 would be a bad fit for their software. Noone is forcing the world to switch immediatly and completly to the GPLv3. What the FSF is doing is offering a license that cannot be used in such devices.
I haven't made up my mind on the issue, but to me it seems that this would make it more difficult to write good open source voting software (for instance) in a world where much of the other software out there is GPLv3'd. A lot of open source is built on other open source --- code is borrowed, libraries are linked, and so on. If the majority of available code was GPL3'd, it'd basically mean not being able to rely on it for these particular tasks.
No doubt code crosses over in both directions frequently, but Debian's had an AMD64 distribution for quite some time now -- it only just missed out being included in the previous Sarge release as stable.
I think the biggest difference between Debian and Ubuntu with AMD64 is that Debian's aiming to make it a pure compiled AMD64 environments, without the library adjustments to allow for x86 binaries as Ubuntu does. In debian, the lib64/ folder is just a sym-link to lib/. There are nice up-sides to this, but I guess the down side is that it's taking longer to get a stable system for some applications that don't like compiling in AMD64.
except OpenOffice, which demonstrates some very dubious programming techniques based around the assumption that the word length and addressing space are exactly 32 bits
FYI, there actually is now an AMD64 port of Openoffice, available in Debian packages (albeit still only in experimental) since about March. I don't rely on it personally just yet because I've found it to be very unstable, but it is getting there. It's linked from http://openoffice.debian.net/, though the actual 2.0.3 files can be found at http://people.debian.org/~rene/openoffice.org/2.0. 3/amd64/.
One episode doesnt give you any reason to be addicted, two gives you something, after three episodes you cant wait to see fourth, after fourth you find yourself thinking how PPG's really work, after fifth you have lost the count.
Taking advantage of the fan base really helps with this. The Lurkers Guide to B5 is quite amazing in the amount of notes for each episode. When I bought the series on DVD, I watched each episode in conjunction with the episode guide (reading the what-to-watch-for bits beforehand, and JMS's catalogued usenet comments about the episode afterwards), and it added a lot to the understanding appreciation, and general enjoyment of the series.
There are quite a few shows toady that've copied the B5 pattern, or built on it in some cases. I have to admit that it's made more of an impression on me than any other SF show, however, just because of how much it revolutionised what could be done for writing and production of TV series' at the time.
No kidding. Good on Techworld for reporting this, but the propagation to Slashdot seems quite redundant. Even if typical slashdot readers didn't know that Debian's had amd64 in its testing distributions for ages, to the point where it only just missed out getting into Sarge, it was already reported in this summary, after all!
I'm a debian user -- I've been a Debian amd64 user for more than a year -- and I like it. (I still wouldn't suggest running AMD64 unless you're prepared to be a beta tester for a variety of desktop applications.) But it's hardly new. As you point out, Ubuntu's been been doing it for a while, too. As with a lot of things, Debian and Ubuntu feed off each other. (Ubuntu gets Debian's package base and stability, Debian gets ports back from Ubuntu adjustments (such as the Openoffice amd64 port, which I think has been primarily Ubuntu-driven).
Well I run 64 bit Debian etch, and it won't let me install 32 bit packages. It's possible to do so in a completely separate userspace, of course. I wouldn't suggest pure 64 bit linux for desktop use at this point, unless you're happy to put up with occasional crashes of desktop applications... especially Office apps. (I'm happy with this myself.)
This is the old "doing it for a living" thing. The enthusiastic amateur can spend as long as they want creating a perfect labour of love, picking and choosing what to work on. The paid pro has a very limited amount of time, can't always choose what to work on and has done it so many times before not everything they do piques their interest 100% any more.
Absolutely -- I agree with you on this point, but I don't think this should detract from the point that media companies aren't actually providing good journalism in many cases. After all, the journalists out there at the moment are a consequence of the people and businesses who hire them. I'd have to admit that this issue is much bigger than just journalism, though. The reason it's a problem is because people put up with it.
When I go home and watch the evening news on TV, for instance, I frequently get the impression that the people on screen couldn't possibly have been hired for their journalism skills as much as for their ability to generate superlatives and cliche's, make something sound dramatic, and generally hold an audience that might switch channels. I agree that most people wouldn't do much better if they were doing it for a living, but I also think it'd be difficult for most of those people to even get such an opportunity. They wouldn't fit the mould of what the media business probably wants.
It's often similar with newspapers and radio, although I at least have some respect for some newspapers, because they're happy to publish people's criticism of them... in a semi-weblog like way. I often flip through letters to the editor -- not because I'm expecting many of them will be worth reading, but because they're a good indication of when something that the newspaper has published is controversial.
I should definitely mention that my perspective is coming from a relatively small country (New Zealand) that has bugger-all decent media. Most of the good journalism I'm familiar with (and plenty of bad) comes from overseas. There's a little good journalism locally, but it mostly falls below the radar. The main exception, which is in a horrible slot of Sunday morning public radio, is a programme called MediaWatch, which evaluates and comments on how the media's been acting in the last week or so.
Only? Since when was it expected that any bloggers were journalists? The only blog I know of that even comes close to journalism is Slashdot, and we all know how that turned out...
It also depends on how you'd define journalism. There are a lot of people I know who have no journalism training, who I'd consider much better journalists than many of the paid front-line journalists for newspapers, TV and radio. There have been more than enough times when I've felt irritated that a journalist didn't actually know (or care) anything about what they were reporting about, at least as much as looking good, being noticed, and being entertaining.
The article itself claims that the 5% figure contradicts perceptions that weblogs are "remaking journalism", but the low figure isn't exactly a surprise for the reasons you and others have been mentioning. Personally I don't think the overall percentage itself isn't anywhere near as relevant as the small number of people who run high quality weblogs that really do provide better quality reporting than many recognised journalists. These weblogs are directly accessible, and usually free, unlike a lot of traditional reporting. The biggest problem I see with weblogs is that it now becomes the reader who has to decide what's worth reading, instead of an editor.
However, to be fair, Microsoft should not be prohibited from being one of those bidders. For example, if Google, Firefox, and Real offer Dell millions to make Google, Firefox, and Real the default search, browser, and media player, respectively, Microsoft should be allowed to outbid Google, Firefox, and/or Real to make Live.com, IE, and WMP the default search, browser, and/or media player.
Isn't this exactly where the logic was behind splitting Microsoft into two companies, before that idea was quashed by political agendas?
If Microsoft Apps (TM) made applications independently and had to play by the same rules as everyone else, it might actually be fair. As it is, Microsoft controls the entire desktop, including the ability to lazily restrict access to certain API's (and tie up the courts in the process). With the cash generated by the dominance of its desktop, Microsoft can afford to out-bid anyone it likes, letting it keep control, albeit at a substantial loss for that particular business area due to bidding amounts that independent competitors couldn't possibly afford.
Personally I wish OEM's would just install good software in the interests of their customers, instead of desktop-and-processor-hogging spyware, but I doubt that'll happen with large OEM's for a variety of reasons.
Since then the programmers have taken over. HTML documents need to have an XML namespace declaration at the top that most mortals can't remember. The CSS inheritance model is nonsensical, I need a 2-page cheat-sheet to get the syntax right, its designer thinks declaring aliases are 'too complex' and it takes a bona fide css expert to get css positioning working across browsers with a design that survives user-preferred fonts.
I'm not an expert on the subject, but I'd thought that one of the fundamental ideas behind the web in the first place was that people shouldn't be needing to write in HTML (or CSS) at all. All of this would be accomplished by tools and generators, which would output clean, standards compliant HTML (and CSS) to be parsed and displayed by some form of web browser.
Perhaps what's needed is more stress on web page generation tools to be standards compliant, and more stress on web page parsing tools to complain about data that's not standards compliant.
I've found.Net API's to be quite well documented, at least for what I use them for. If I was only developing native.Net applications that never interacted with anything but.Net interfaces, it'd be quite nice. My own problems have come up when interacting with non-.Net API's, which tend to involve unmanaged memory, and I've been quite surprised how many Microsoft products still seem to rely on them given MS's push for everyone to use.Net.
For me, the Office API has often been really hard to use effectively. It's quite buggy, and the documentation seems to be sparse. There's lots of documentation-by-example for doing specific things, but little specification of what the different bits of the object model actually represent or how they're supposed to act.
The only reasonable solution to this is that most countries make stuff for themselves. India, China and Africa should concentrate of getting the skills and making the stuff they need for themselves, instead of getting skills and making stuff we need, and then shipping all that stuff around the world at great cost to the environment.
I'm not sure I agree with this at all. If you cut trade and make every country isolationist, it reduces the standard of living for everyone. Particularly in smaller countries, there are some things needed for a certain quality of living that simply can't be made because the economy won't support them. They have to be imported, and those countries have to find other things to export in order to pay for them. It's not always material goods that require shipping, either. It's just as often going to be expertise or experience that's traded.
Some countries do have a lower cost of living, completely fairly, and so they're cheaper to produce in, but (I think) this is usually the case because those countries have been cut off in the first place by those with bigger economies (eg. the USA). The costs of living would tend to even out if trade were completely free. Realistically, living in New Zealand, I know that my cost of living would definitely go up if the USA didn't have all the tarrifs on imports of agricultural goods to protect the farmers who are producing inefficiently -- it would mean there would be less goods to go around here, prices here would go up. It'd also let cheaper goods into the USA, and prices would go down. Furthermore, it would mean there would be more money in our economy, I'd get paid more (overall), and I'd actually be closer to being able to afford goods produced in the USA that are made more efficiently than what we can do locally.
On the other hand, some countries are simply cheaper to produce in because they don't have the same human rights restrictions that more developed countries have, and this is the main problem that I have with trade between other countries. Personally I don't think these countries should be traded with until they meet a general standard that's acceptable. It's entirely hypocritical, along the lines of saying "I believe it's wrong to treat people in such a way, but the guy over there doesn't, so I'll pay him to treat his own people unfairly for my benefit." As it is, New Zealand's government is pursuing a free trade agreement with China, primarily because the USA has cut them off due to some unrelated disagreement about nuclear arms, not to mention a bunch of riders that the USA tries to hook onto any trade deals it allows. Whatever.
Trade really does benefit everyone, but only when everyone's playing by the same rules, and at the moment this isn't the case. Kurt Vonnegut has a great quote in Player Piano which I don't remember exactly, but it's along the lines of "If you compete with slaves, you become slaves".
Right there is why Microsoft is the most successful software company in the world -- respect for developers.
Perhaps I've had different experiences from you, but I completely disagree with this statement and your comments that followed. Lately I've been developing in-house applications, particularly ones intergrated with Office applications, and two of the most irritating problems we've had are directly related to bugs in Microsoft's API's that have existed for years. In one example, Word randomly fails to load plugin and template code for no apparent reason, causing unpredictible problems later on. In another example, Outlook events don't fire properly, preventing unmanaged code from cleaning up cleanly (yes, 5 years after the great managed.Net framework was released, Office still uses hacks to integrate it).
Overall though, this isn't the real problem. Most software has bugs at some point. The biggest problem I have with Microsoft and its relation with developers is that the company has invented a new level of obscurity in open bug management. If you pay thousands of dollars for a Microsoft support contract, it's sometimes possible to get attention from Microsoft, and a bug might be reported internally if someone decides it's worth it. It's just as likely that you'll be given a workaround that ignores the real problem and just makes your own code harder to maintain. Most problematically, there's no real open-ness about the process at all for anyone stuck on the outside, and trying to find out the current status of possible bugs is just as annoying. I've lost count of the number of times I've received an obscure error message, and a quick search of Usenet has revealed that hundreds of others have reported seeing the same thing: nobody really knows what it means, nobody has a fix, and nobody has a clue if Microsoft knows there's a problem.
I also dislike the way that Microsoft uses its products to drag developers and systems integraters, kicking and flailing, on whatever path it's decided to follow. Want to get all old.Net v1 apps recompiled in.Net v2? You can't. Want to support PC's that don't have.Net v2 installed and still use VS2005? You can't. This is a real pain, because we've found that in an environment where.Net code integrates with MS API's (MSIE in this case), having.Net v1 and.Net v2 installed on the same PC makes things very unstable, and it's a pain trying to get tird party libraries updated to the new environment. Yes, open source projects do the same -- usually on a smaller scale -- but other projects at least have an open development process that makes it much easier for developers to find out what's changed, why their code isn't working, what they need to do to update their code, and to easily search for and report bugs if appropriate.
Even though I develop on a Windows platform, I make a conscious effort to avoid locking the development into Microsoft tools. eg. Even though SourceSafe has improved quite radically in the past year or so, I don't want to lock our code into a SourceSafe repository. I use Subversion instead, because I have a lot more confidence that Subversion will continue to run, independent of the OS/version running underneath it. I also have a lot more confidence that I won't end up having to upgrade or re-write a bunch of random other applications simply because Subversion needs an update.
If Microsoft would just shut up with all the market-babble and open up their development process in ways that would allow people to directly contact developers, report and comment on bugs, and allow others to reliably support their products, things would be so much easier. I doubt this is ever going to happen, though, primarily because it'd seriously cut into Microsoft's control of the market and main source of income.
Slashdot Mirror
Textbooks are expensive, but in my experience the majority of books that need to tell people how many planets are in the solar system are not textbooks. People who own astronomy textbooks already know how many planets there are, and most are likely to be more interested in the fact that Pluto is a small icy double-body in the outer solar system, than in whether people classify it as a planet. If anything, the new classification will make it easier for textbook users, because it more cleanly groups the different types of bodies in the solar system... but any good textbook probably attended to that already.
Books likely to need updating are common library books, and encyclopedias (which need updating frequently for all the other information they contain, anyway). In either case, the types of people who own and use these books aren't the types of people who would care a lot if it was out of date. If this decision has any influence on book sales (textbook or otherwise) I'd be very surprised, and I'm sure the book publishers know this too.
I'm starting to agree with this too, but it's taken a while of thinking to get to this stage. The main issue I've been trying to work out, which is one that's come up in previous GPL3 discussions, is with open source voting machines, and the legality of using GPL3'd code to write software for open source voting machines that might need to be restricted to running a specific revision of specific software to be trustworthy. In that case it's really in everybody's interests that the code can't simply be changed by someone at the end of the line who controls a voting booth, and I'd like to think that secure-from-tampering hardware is still possible with GPL3.
I think the best answer to this, if it holds up legally, is that the central voting authority which owns the machines (state authority, national authority, whatever), is the authority that should be adjusting the hardware to restrict it to running a particular set of code. That said, I'm still not sure exactly how the legal side of it works out, which would probably depend on a lot of details about who owns what, who's employing who, who's giving what to who, and what counts as "distribution".
It'd be nice if Google would eventually offer an encryption system such that no plain-text documents ever saw Google's servers. (ie. It'd need to be encrypted in the client before being saved, and ideally the whole system for this would have verifiable source available.) Collaboration could be a bit of an issue, but it'd still be feasible if the people collaborating on the document could share a common key.
The biggest down side for Google in this scenario is that it wouldn't (easily) be able to scan documents for any future targeted advertising, or whatever else they might have in mind.
All this said, GMail (let alone all the other webmail out there) has already demonstrated that there are massive amounts of people out there who are happy to put personal and sensitive information on someone else's servers. (Ironically for me, I'm also one of them in the case of gmail, I guess.)
I know this is a nitpick -- I fully agree that clients often don't appreciate how serious some bugs are, but I don't think it'd be unreasonable for a client to want to prioritise the look of a report over making the query 10% faster. Really it'd come down to how critical the exact-ness of the template is. If it's a government organisation, for instance, there might be legislation in place that says it has to be a certain format or a law's being broken. Or it might just be an eyesore to look at. You really can't know without talking to the customer or reading the specs (if they've been well authored).
If the query was returning the wrong data, or if it took orders of magnitude longer to run than it needed to, it'd definitely be worth looking at first as a serious bug. Really though, if it's a report that a user has to sit and wait for anyway, 10% longer running time behind their other tasks probably won't have nearly as much impact on their work day as having to use a report that looks different from how they need it to look. If it's a report that runs nearly instantaneously, 10% difference is nothing anyway.
I think this is an important point about open source in general, and it reaches much further than just drivers. It applies to the entire collection of applications in a distribution.
This is one of the key things that I think Open Source Software and licensing has going for it. On the surface, it allows someone to take the code and use it elsewhere, but the fundamental nature of the licenses also means that support for existing software can be transferred much more easily. This makes it a lot simpler to have a decent and reliable package management system in open source OS distributions. Essentially, it makes it possible for a single entity to support an entire system, and have the freedom to update any part of it. Even though it sometimes means minor forking of code to fit with a particular distro, it's already been shown, for a long time, that this can be done in a maintainable way. Distros can successfully tinker with code for their own distro while still keeping it in sync with upstream releases.
People often complain about linux app installation not being as simple as download, click, run, but I can't imagine how it'd be possible to have anything like yum or apt-get, let alone a stable underlying package management system like rpm or dpkg, in a Windows environment with closed source licenses. Windows has a built-in package manager, but it's very basic compared with others. Personally I've always had problems trusting that installers are doing what I hope it's doing... keeping in mind that most application installers tend to come from third parties who might or might not know what they're doing. When the installation package comes via a central provider who's certified it for use with their system, it's much more trustworthy.
It's simply not possible for Microsoft, or any other organisation, to collect together thousands of third party Windows applications and provide them in a stable, integrated repository that's supported by a single provider. The closed source licenses of software vendors typically prevent this, and they also typically prevent anyone from tinkering with source code and/or re-releasing the product, if source code is even available at all.
At best, Microsoft can provide some level of stability with Microsoft provided applications (Office, IE, some specific dev tools, etc), but after that the user is on their own, and has to get support from elsewhere. To top it off, Microsoft's own closed source policies for Windows means that third party supporters don't have full control to keep their apps stable, either. (eg. Any Windows bugs that affect an application can only be fixed at the whim of Microsoft, fixes might break other apps that a user needs, and so on.) As soon as a user wants to use something that Microsoft doesn't provide, or wants to do something in a way other than the one and only Microsoft way, it's necessary to go to a third party, and then it risks becoming a support nightmare.
I use open source software for all sorts of reasons, but I think most of them trace back to this basic thing that the licenses of individual bits of software allow for my entire system to be much more stable. I use two binary blob drivers on my laptop (nvidia and broadcom) because I have to to get it to work how I want it to, but the next laptop I buy will definitely be scrutinised much more for its hardware -- both drivers have just been such a hassle to maintain in a way that keeps the system stable.
I agree. The key for free software (to me) is not whether the developer(s) are engaging with a community to develop the software -- it's whether people in the community can freely fork the code and continue to develop it independently if they decide they don't like the way that the original developer(s) are going.
As long as Open Solaris is licensed in such a way, it's fine by me to refer to it open source.
More likely it's just one part of the company not being in-tune with another part of the company. I'm sure there are lots of people at IBM who'd understand the main points of OSS. It's a shame that this got as far as an executive's press release, though. It implies that the people at the top of IBM's open source project don't really get it, even if others do.
Perhaps. Whether this is true or not, I'm certain that Asimov himself is on record as saying (probably in one of his intros) that the Three Laws are primarily a story device for interesting fiction. When he got started writing robot stories, he was most interested in writing stories that weren't about robotic terror machines going crazy. Many people were afraid of and felt threatened by robots in the 1950's.
Developing the Three Laws of Robotics and treating them as un-breakable was a convenient way to put boundaries on things (without having to explain their implementation). It made it easier to justify his robots acting the way they did, creating new and interesting stories as a result. By the end of it, Asimov was also on record as saying something along the lines of that he thought he'd probably explored all the possibilities of how robots might react in different situations if they were constrained by these laws.
Asimov was probably one of the earliest people to recognise that robots and machines could be beneficial to humanity, and didn't need to replace humanity. He got his message across by convincing people that they could be built with the primary intention of helping people. Trying to apply the laws to reality, however, is asking for trouble.
How long until Slashdot can get a category for articles by Dvorak? I'd really like the opportunity to be able to filter his stories from my front page.
I realise that your complaint is mostly about the lying aspect and that you probably weren't meaning to imply that propaganda is illegal. Assuming there are lies intended to demean someone without justification then by all means let it go through the courts, but I think it's worth making it clear that propaganda doesn't imply lying or illegality, paid or not. In fact, the word itself means little more than the spreading of views and interests that promote a particular cause, influence people's opinions, and whatever else.
I haven't made up my mind on this issue. Part of me wants to agree that democracy should include (by definition) free access to reliable and objective information about important issues that isn't influenced by propaganda, but realistically that information still is available to anyone who chooses to look for it. On the other hand, people pushing their ideas and trying to influence opinion is a side effect of free speech -- democracy invites it as much as something like communism probably invites government official corruption. I'm not sure how it can really be changed without some dramatic changes to the way government works. Personally I'm fortunate to live in a relatively small country, and I get the impression that the result is a national government which is much more open and less like a dinosaur than the US Federal Government. Maybe it's just that democracy doesn't scale well, or that the US form of democracy hasn't scaled well.
We definitely care about it at my workplace, because it's likely that the next desktop we roll out to all our users will be a Vista desktop. We're not exactly going to roll it out immediately on release, but we do want a good idea of how long the final product will have been out there and tested before we use it on a large scale ourselves.
A 3 or 4 month delay in the Vista release probably means we'll have to factor in a 3 or 4 month delay in our own rollout schedule, just so we can have enough time to play with it properly (post-beta), check everything works, wait for other vendors to sort out their apps to fit Vista, and sort out problems with our specialised apps. This affects all kinds of things in the running of the business, and on short notice a delay can be a real annoyance. (One that we're prepared for given Microsoft's repution, but an annoyance all the same.)
This is true, and good for them for reaching an agreement, although I'm curious to know what the hooks are in the agreement since Real would probably have been able to take and re-package Firefox regardless given its licensing. The main thing that I'm wondering about this is how much and in what ways Real might plan to modify Firefox for their own packaged version.
Will people with the Real build of Firefox get selective popup advertising from Real.Com? Will automatic upgrades come from the official Firefox distribution site, or from Real.Com? Will people easily be able to revert their installation to a generic Firefox? Given Real's reputation and past actions, it wouldn't surprise me too much if this became an issue.
I don't know about you, but I don't subscribe to either of these.
I consider myself part of the Open Source community because I both use Open Source, and from time to time I've also written and released my own Open Source. I don't particularly care about spreading the philosophy (although I'm happy to explain it to people), and I don't feel the need to make people use it (although I'm happy to help them if they want to, within reason).
Personally I like and use Open Source software because in the ways that I like using software, I find it to be of superior quality and better suited to my needs for a variety of reasons. Running campaigns and trying to convert people to new philosophies has nothing to do with it.
Individual people or organisations within the open source community might have goals, but I don't think it's a serious problem if different groups disagree. I'm also not sure if it's meaningful to claim that people should be aiming for a goal just because they're involved in open source. If anything, perhaps one issue that could be addressed is how to better identify different interest groups without trying to bundle them all into the "Open Source Software Community" basket.
I haven't made up my mind on the issue, but to me it seems that this would make it more difficult to write good open source voting software (for instance) in a world where much of the other software out there is GPLv3'd. A lot of open source is built on other open source --- code is borrowed, libraries are linked, and so on. If the majority of available code was GPL3'd, it'd basically mean not being able to rely on it for these particular tasks.
There's always the BSD-licensed code, of course.
No doubt code crosses over in both directions frequently, but Debian's had an AMD64 distribution for quite some time now -- it only just missed out being included in the previous Sarge release as stable.
I think the biggest difference between Debian and Ubuntu with AMD64 is that Debian's aiming to make it a pure compiled AMD64 environments, without the library adjustments to allow for x86 binaries as Ubuntu does. In debian, the lib64/ folder is just a sym-link to lib/. There are nice up-sides to this, but I guess the down side is that it's taking longer to get a stable system for some applications that don't like compiling in AMD64.
FYI, there actually is now an AMD64 port of Openoffice, available in Debian packages (albeit still only in experimental) since about March. I don't rely on it personally just yet because I've found it to be very unstable, but it is getting there. It's linked from http://openoffice.debian.net/, though the actual 2.0.3 files can be found at http://people.debian.org/~rene/openoffice.org/2.0. 3/amd64/.
Taking advantage of the fan base really helps with this. The Lurkers Guide to B5 is quite amazing in the amount of notes for each episode. When I bought the series on DVD, I watched each episode in conjunction with the episode guide (reading the what-to-watch-for bits beforehand, and JMS's catalogued usenet comments about the episode afterwards), and it added a lot to the understanding appreciation, and general enjoyment of the series.
There are quite a few shows toady that've copied the B5 pattern, or built on it in some cases. I have to admit that it's made more of an impression on me than any other SF show, however, just because of how much it revolutionised what could be done for writing and production of TV series' at the time.
No kidding. Good on Techworld for reporting this, but the propagation to Slashdot seems quite redundant. Even if typical slashdot readers didn't know that Debian's had amd64 in its testing distributions for ages, to the point where it only just missed out getting into Sarge, it was already reported in this summary, after all!
I'm a debian user -- I've been a Debian amd64 user for more than a year -- and I like it. (I still wouldn't suggest running AMD64 unless you're prepared to be a beta tester for a variety of desktop applications.) But it's hardly new. As you point out, Ubuntu's been been doing it for a while, too. As with a lot of things, Debian and Ubuntu feed off each other. (Ubuntu gets Debian's package base and stability, Debian gets ports back from Ubuntu adjustments (such as the Openoffice amd64 port, which I think has been primarily Ubuntu-driven).
Well I run 64 bit Debian etch, and it won't let me install 32 bit packages. It's possible to do so in a completely separate userspace, of course. I wouldn't suggest pure 64 bit linux for desktop use at this point, unless you're happy to put up with occasional crashes of desktop applications... especially Office apps. (I'm happy with this myself.)
Absolutely -- I agree with you on this point, but I don't think this should detract from the point that media companies aren't actually providing good journalism in many cases. After all, the journalists out there at the moment are a consequence of the people and businesses who hire them. I'd have to admit that this issue is much bigger than just journalism, though. The reason it's a problem is because people put up with it.
When I go home and watch the evening news on TV, for instance, I frequently get the impression that the people on screen couldn't possibly have been hired for their journalism skills as much as for their ability to generate superlatives and cliche's, make something sound dramatic, and generally hold an audience that might switch channels. I agree that most people wouldn't do much better if they were doing it for a living, but I also think it'd be difficult for most of those people to even get such an opportunity. They wouldn't fit the mould of what the media business probably wants.
It's often similar with newspapers and radio, although I at least have some respect for some newspapers, because they're happy to publish people's criticism of them... in a semi-weblog like way. I often flip through letters to the editor -- not because I'm expecting many of them will be worth reading, but because they're a good indication of when something that the newspaper has published is controversial.
I should definitely mention that my perspective is coming from a relatively small country (New Zealand) that has bugger-all decent media. Most of the good journalism I'm familiar with (and plenty of bad) comes from overseas. There's a little good journalism locally, but it mostly falls below the radar. The main exception, which is in a horrible slot of Sunday morning public radio, is a programme called MediaWatch, which evaluates and comments on how the media's been acting in the last week or so.
It also depends on how you'd define journalism. There are a lot of people I know who have no journalism training, who I'd consider much better journalists than many of the paid front-line journalists for newspapers, TV and radio. There have been more than enough times when I've felt irritated that a journalist didn't actually know (or care) anything about what they were reporting about, at least as much as looking good, being noticed, and being entertaining.
The article itself claims that the 5% figure contradicts perceptions that weblogs are "remaking journalism", but the low figure isn't exactly a surprise for the reasons you and others have been mentioning. Personally I don't think the overall percentage itself isn't anywhere near as relevant as the small number of people who run high quality weblogs that really do provide better quality reporting than many recognised journalists. These weblogs are directly accessible, and usually free, unlike a lot of traditional reporting. The biggest problem I see with weblogs is that it now becomes the reader who has to decide what's worth reading, instead of an editor.
Isn't this exactly where the logic was behind splitting Microsoft into two companies, before that idea was quashed by political agendas?
If Microsoft Apps (TM) made applications independently and had to play by the same rules as everyone else, it might actually be fair. As it is, Microsoft controls the entire desktop, including the ability to lazily restrict access to certain API's (and tie up the courts in the process). With the cash generated by the dominance of its desktop, Microsoft can afford to out-bid anyone it likes, letting it keep control, albeit at a substantial loss for that particular business area due to bidding amounts that independent competitors couldn't possibly afford.
Personally I wish OEM's would just install good software in the interests of their customers, instead of desktop-and-processor-hogging spyware, but I doubt that'll happen with large OEM's for a variety of reasons.
I'm not an expert on the subject, but I'd thought that one of the fundamental ideas behind the web in the first place was that people shouldn't be needing to write in HTML (or CSS) at all. All of this would be accomplished by tools and generators, which would output clean, standards compliant HTML (and CSS) to be parsed and displayed by some form of web browser.
Perhaps what's needed is more stress on web page generation tools to be standards compliant, and more stress on web page parsing tools to complain about data that's not standards compliant.
I've found .Net API's to be quite well documented, at least for what I use them for. If I was only developing native .Net applications that never interacted with anything but .Net interfaces, it'd be quite nice. My own problems have come up when interacting with non-.Net API's, which tend to involve unmanaged memory, and I've been quite surprised how many Microsoft products still seem to rely on them given MS's push for everyone to use .Net.
For me, the Office API has often been really hard to use effectively. It's quite buggy, and the documentation seems to be sparse. There's lots of documentation-by-example for doing specific things, but little specification of what the different bits of the object model actually represent or how they're supposed to act.
I'm not sure I agree with this at all. If you cut trade and make every country isolationist, it reduces the standard of living for everyone. Particularly in smaller countries, there are some things needed for a certain quality of living that simply can't be made because the economy won't support them. They have to be imported, and those countries have to find other things to export in order to pay for them. It's not always material goods that require shipping, either. It's just as often going to be expertise or experience that's traded.
Some countries do have a lower cost of living, completely fairly, and so they're cheaper to produce in, but (I think) this is usually the case because those countries have been cut off in the first place by those with bigger economies (eg. the USA). The costs of living would tend to even out if trade were completely free. Realistically, living in New Zealand, I know that my cost of living would definitely go up if the USA didn't have all the tarrifs on imports of agricultural goods to protect the farmers who are producing inefficiently -- it would mean there would be less goods to go around here, prices here would go up. It'd also let cheaper goods into the USA, and prices would go down. Furthermore, it would mean there would be more money in our economy, I'd get paid more (overall), and I'd actually be closer to being able to afford goods produced in the USA that are made more efficiently than what we can do locally.
On the other hand, some countries are simply cheaper to produce in because they don't have the same human rights restrictions that more developed countries have, and this is the main problem that I have with trade between other countries. Personally I don't think these countries should be traded with until they meet a general standard that's acceptable. It's entirely hypocritical, along the lines of saying "I believe it's wrong to treat people in such a way, but the guy over there doesn't, so I'll pay him to treat his own people unfairly for my benefit." As it is, New Zealand's government is pursuing a free trade agreement with China, primarily because the USA has cut them off due to some unrelated disagreement about nuclear arms, not to mention a bunch of riders that the USA tries to hook onto any trade deals it allows. Whatever.
Trade really does benefit everyone, but only when everyone's playing by the same rules, and at the moment this isn't the case. Kurt Vonnegut has a great quote in Player Piano which I don't remember exactly, but it's along the lines of "If you compete with slaves, you become slaves".
Perhaps I've had different experiences from you, but I completely disagree with this statement and your comments that followed. Lately I've been developing in-house applications, particularly ones intergrated with Office applications, and two of the most irritating problems we've had are directly related to bugs in Microsoft's API's that have existed for years. In one example, Word randomly fails to load plugin and template code for no apparent reason, causing unpredictible problems later on. In another example, Outlook events don't fire properly, preventing unmanaged code from cleaning up cleanly (yes, 5 years after the great managed .Net framework was released, Office still uses hacks to integrate it).
Overall though, this isn't the real problem. Most software has bugs at some point. The biggest problem I have with Microsoft and its relation with developers is that the company has invented a new level of obscurity in open bug management. If you pay thousands of dollars for a Microsoft support contract, it's sometimes possible to get attention from Microsoft, and a bug might be reported internally if someone decides it's worth it. It's just as likely that you'll be given a workaround that ignores the real problem and just makes your own code harder to maintain. Most problematically, there's no real open-ness about the process at all for anyone stuck on the outside, and trying to find out the current status of possible bugs is just as annoying. I've lost count of the number of times I've received an obscure error message, and a quick search of Usenet has revealed that hundreds of others have reported seeing the same thing: nobody really knows what it means, nobody has a fix, and nobody has a clue if Microsoft knows there's a problem.
I also dislike the way that Microsoft uses its products to drag developers and systems integraters, kicking and flailing, on whatever path it's decided to follow. Want to get all old .Net v1 apps recompiled in .Net v2? You can't. Want to support PC's that don't have .Net v2 installed and still use VS2005? You can't. This is a real pain, because we've found that in an environment where .Net code integrates with MS API's (MSIE in this case), having .Net v1 and .Net v2 installed on the same PC makes things very unstable, and it's a pain trying to get tird party libraries updated to the new environment. Yes, open source projects do the same -- usually on a smaller scale -- but other projects at least have an open development process that makes it much easier for developers to find out what's changed, why their code isn't working, what they need to do to update their code, and to easily search for and report bugs if appropriate.
Even though I develop on a Windows platform, I make a conscious effort to avoid locking the development into Microsoft tools. eg. Even though SourceSafe has improved quite radically in the past year or so, I don't want to lock our code into a SourceSafe repository. I use Subversion instead, because I have a lot more confidence that Subversion will continue to run, independent of the OS/version running underneath it. I also have a lot more confidence that I won't end up having to upgrade or re-write a bunch of random other applications simply because Subversion needs an update.
If Microsoft would just shut up with all the market-babble and open up their development process in ways that would allow people to directly contact developers, report and comment on bugs, and allow others to reliably support their products, things would be so much easier. I doubt this is ever going to happen, though, primarily because it'd seriously cut into Microsoft's control of the market and main source of income.