Not all OSS is high quality, far from it. And last but not least, not all of it is maintained on a decently regular basis. I know a lot of OSS projects, some of which quite good, which have gone unmaintained, or are maintained once in a blue moon - that is unprofessional. And that's the very nature of OSS
It sounds like the nature of nearly all software to me. There are a few projects, open source and closed source, that are of high quality, well and consistently maintained. The majority aren't, and I don't think their philosophy has much to do with it.
The big (long term) advantage I see with open source is that if the developers decide to ditch it, you always have the option of paying someone else to keep it maintained for you... or doing it yourself if you have the expertise. If closed source developers decide to kill their project, on the other hand, that option isn't at all certain. There are plenty of closed source software projects out there that are no longer maintained.
Some journalists are just stuck in an awful system
on
Bad Science in the Press
·
· Score: 2, Interesting
Journalists exist to be published. That is their function -- that's what they love, to see their name in print. They don't really care what they say exactly; they only care that their article pleases their editors, which in turn sells more newspapers or magazines.
I think your problem is at least as much with the editors as with the journalists. I realise what you're saying, but journalists aren't all like that. I have a good friend who's an amateur astronomer, and she's also a freelance science writer. She has things published in daily newspapers and/or weekly magazines about every week or two. She gets paid a commission, but definitely doen't make enough to cover the time she spends on it. Her primary motivation for doing it is to simply try and get some science writing in the media at all.
The irony is that even though she knows exactly what she's talking about when writing something, and she puts a huge amount of effort into being careful and accurate about what she writes, she still has to fight with editors. Probably most of the articles she writes end up getting lazily re-written in some way, even if it's just re-typed for some reason with half the words messed up. (eg. In an article last week, they changed "mass spectrometer" to "massive spectrometer", which is completely different!) It's not uncommon for chunks of writing to be cut without proper consultation, and with no respect for the context, or how they might be changing the message of the article. On a couple of occasions, they've held back time-critical articles and published them weeks after they were actually relevant, as if they didn't even bother to check the content properly.
It's not just the editors in this case, either. It's the whole system that involves deadlines and priorities that the people in the business give themselves. Editors of regular media publications just don't get to be editors by knowing about or having much respect for science. In the cases I've mentioned above, they've made a broad descision to step up and help to publicise science, but in reality they don't really care too much about the specifics of what they're doing -- it's for show as much as anything.
Some journalists "exist to be published", but I think the main ones in that frame are the staff journalists who are being paid a salary. Those people are only a subset of all the writing that you're likely to see in many publications. The problems are likely to come especially at times when they're being asked to write about things they don't find particularly interesting... and there simply aren't many staff journalists out there who find scientific topics anywhere near as interesting as things like politics, crime or business, for instance.
We've also had one "it'll never happen" situation where I did actually give in and do it the way the Board specified (dynamic content served by ASP.NET instead of Perl, on a server too old to support ASP.NET reliably).
Did you consider using Mono for hosting of.NET on a non-Windows platform? From the tone of your post, I'm guessing that your boss would have blocked it, but I'm just curious. I'm only just starting to look at Mono (just for fun right now), and I'm wondering how reliable it is compared with Microsoft's implementation.
No, the question is, why would you put voice-over-ip into a word processing document? The purpose of a word-processing document (text of, e.g., laws and regulations) is entirely different from the purpose of a video or audio record.
Because the way Microsoft's going, they don't see it as a Word Processing document any more. They see it as an access point for exchanging information, and there's really no reason (in Microsoft's view) why there should be boundaries on what types of information or interaction should be found in what places.
I don't particularly like the way that Microsoft operates, and I have no doubt that their criticism of the Massachusetts decision is plain and simple FUD. As you say, it probably makes much more sense for a government entity, and probably most other entities, to use well defined, straight-forward formats for storing its important information. I think it's a bit short-sighted, though, to criticise Microsoft too much for mixing together VOIP and Word Processing.
They're just working with a different model from the traditional metaphor that most people are familiar with. (No, I don't think they're doing a very good job being clear about what metaphor, if any, they are using.) Really though, the only reason people presently think it's strange to mix Word Processing and Voice Over IP is because we've so far been conditioned to think of a Word Processing document as a printable piece of paper. There are plenty of other things with writing on them that aren't paper, and it's not necessarily true that a word processor shouldn't mimic or act like them just as much, if a user wants to use it that way.
In Microsoft's world, someone will be able to click an icon (or access a "thing", or whatever) without having to care about what it is. They'll simply get whatever it contains, whether it's printable typewriter-style text, or some kind of portal for speaking to someone else. Similarly, people will be able to construct whatever media they want, without necessarily needing to think about whether they're dealing with a word processor, a spreadsheet, or a video camera. The biggest barrier is people's ability to adapt from how they relate to things in the real world. Metaphors are really only needed for people who have prior experience with something else, and there are some valid arguments that metaphors cause more problems than they solve.
I don't know if this is an ideal way to go or not, but I think you really need to recognise that Microsoft is not necessarily limiting themselves to the traditional ways of using an Office Suite any more. For better or worse, this is one of the differences between Microsoft and Open Office.
The country is too damn big to have disaster plans for every region, which is why it's the responsibility of local and state governments to have plans. Why the hell didn't the mayor of New Orleans have a plan to get HIS OWN CITIZENS out?? The guy is being a total a-hole blaiming the feds for his own failure, as well as the failure of the governor of Louisiana. [...] The feds are not designed to move fast, combined with the fact that it was an incredible mess
The Feds aren't designed to move fast, but they really should have thought ahead so that they'd never have to.
I'm not a citizen of the USA. If I was, though, I wouldn't think it too unreasonable for the Federal Government to at least coordinate with the state governments, make sure they're prepared, and to help arrange preparations if necessary. This might easily involve helping to provide caches of suitable resources in strategic places (whether in the state or nearby), providing expertise to make sure that the response plans are feasible, coordinating agreements with neighbouring states, etc. It might also involve issuing warnings directly to citizens if the feds believe that state officials aren't doing their job properly, or are ignoring federal advice.
If the Federal Government had had a proper programme ages ago to ensure that state governments had identified, were aware of, and properly prepared for possible future disasters in their region (such as this one), it might never have been such a serious problem.
For what it's worth, Microsoft is (also) a New Zealand company. It was registered at the NZ Companies Office on 29 May 1991. It really is an American company, but I guess they at least have a claim to comment on New Zealand law.
I think one of the problems with this method at the moment is that most organisations don't provide clear information about exactly what someone can expect by checking the box. Often it's buried somewhere deep within the privacy policy, but it's not exactly obvious.
Before I check such a box, I like to be confident that I understand basic things like:
How frequently I'm likely to receive mail.
What type of content it will contain. For me, there's a big difference between something like:
"Try our cheap trip to Hawaii", and
"We've reduced the cost for upgrading to business class. Would you like to do this for the trip that we have you on record as having booked a month ago?"
How easy it's going to be to get off the list. Ideally I want to know how to get off the list before signing up to it. I also want to have some clear contact information for someone who I can contact if their unsubscription system breaks.
Who's actually going to be mailing me. Often businesses like to be able to send emails for their "business partners", but I want to know what this means. If it means that they're going to send me any old spam that someone pays them to send me (which is often the case). Basically you go down in their book as an asset after ticking the box, because they can make money off other people by sending you email. I normally won't check the box if there's any doubt about it. But it might mean that every so often, there actually are things that they think I might find useful, and they might want to let me know because of that rather than because someone's paying them. If this is the case, and I trust them, then I might consider signing up.
Most boxes don't actually do this. They just say inane things like "Click here to receive great deals from us and our partners in your email." I'd rather they said something like:
"Click here and we'll keep you informed about deals we have in the future.
"For examples of what you'll receive, check out [this list of some of our past deals]. We'll send about one email a week, and you stop us from sending them whenever you like. ([Click here for more info about how this works.]) We might also send you information from other businesses if we think it's of use to you, but we won't be give them your contact details (without specific permission), or accept money from them to forwarding it."
I guess it's a bit more verbose, but to me it's a hell of a lot clearer and more trustworthy. Then again, I realise that most people don't seem to think/care about this type of thing as much as I do. I'm sure I'm not the main target of many marketing people... I just get annoyed as collateral damage.
Renting software in the 90's
on
MS & Game Rentals
·
· Score: 2, Insightful
I remember the '90's. I remember that it was actually legal for people and businesses to lend or rent, or even re-sell the software they'd purchased to other people. The only condition was that it wasn't allowed to be run in more than one place at a time. Locally, we even had rent-by-mail companies that would take out full page advertisements in magazines, and post you software to use for a limited amount of time before you were (legally) required to uninstall it and return it.
Software companies -- not even Microsoft, but especially Microsoft -- went to great lengths to inform their customers that they were allowed to install Microsoft Word from their work onto their home PC, as long as it was only being used on one PC at a time. After all, it was clear that the software was licenced for use to the person who'd paid for it, so they were allowed to use it wherever it was most convenient for them without having to pay multiple times.
Some software had basic copy protection (eg. flight simulator games that asked you to quote a random word from page 215 of the manual), but there was rarely serious DRM. Software companies were quite clearly concerned about software piracy, but they weren't trying to wipe out customer's existing rights to solve that problem. On the contrary, most software companies and consumers actually acted as if they "trusted" each other, compared with today.
Somehow, this whole attitude has been lost in recent years, at least in commercial software. A major part of it seems to have been about when the media barons leapt to digital and brought all their annoying views on draconian copy protection with them. There's no way in hell you'd be allowed to casually install a work copy of MS Word on a home PC today, unless your company had a special agreement that they'd paid extra for. Even with this, there would still be DRM loopholes to jump through. You'd probably end up in jail for 15 years as an example to other "criminals" if the Microsoft-sponsored BSA lawyers had their way.
I coincidentally watched Dark City on video a couple of weeks before The Matrix was released. I don't remember either of them very well any more, but I do remember thinking, as I was watching it, that the Matrix seemed like a rip-off of Dark City in many ways. It was probably just coincidence in hindsight.
Of course, as others have also pointed out, Dark City itself was preceeded by Metropolis.
I think the article sums up this area quite nicely:
Arguably, one of the worst-case scenarios is the so-called "colorable case" - where there is no substance to an IP claim, but a company is forced to waste millions of dollars to defend the claim or settles early for a large sum to make the case go away. The average US patent action is estimated to cost $2m, according to the American Intellectual Property Lawyers' Association.
I'm glad the author included this note. It indicates quite clearly why a lot of companies see it as being so important to be cautious of lawsuits. It's not the possibility of losing that matters, it's the cost of fighting. If the cost of fighting is more than a company can afford, it just make sense to settle, and then the overly litigous company wins. Even enlightened CEO's often want indemnification--not because they think a company like SCO has a hope of winning a lawsuit, but because they're concerned about how much it might cost if they're even targeted.
Personally I think there need to be some changes to the system so that those who abuse it in this way are penalised much more, and also more quickly. The fact that SCO was both able to be so noticed in the first place, and can even continue dragging its corpse around today to threaten people with its stench without the likelihood of serious charges being brought against it, its executives or its legal team, suggests that not quite enough measures yet exist.
Even "generating better results" sounds like a commercial. Better results for who? Better results for me would be fewer ads, in every medium.
As much as I agree with the general sentiment of this thread about marketing (which I frequently get sick of), there are times where I actually appreciate it. I don't like marketing and advertising that's in my face, and I don't like marketing that lies to me. But some marketing material is just out there to be informative for people who want it and ask for it, without being in anyone's face.
Better results for me means being fully informed about all the relevant options I have, at a time and level of detail of my choosing. This is also a type of marketing, and it's one that I respect. I do know some marketing people who focus on this goal, and I appreciate it.
I actually like the way that the shoe salesman walks up to sell me a shoe when I walk in. I really have no idea what I want and it's not a decision I want to make. What I care about is trusting the guy to sell me something that works, and that's what will get me to come back again and again. That's also marketing. The guy's job relies on him selling lots of shoes, but he knows that his best approach is just to be honest with people. (and to chat, and joke, and so on.)
I also quite like the way that Amazon suggests books for me to read. It's only there when I ask for it, it often offers good suggestions, and every so often they might get a sale out of it. That's the type of marketing that I like.
Thanks for the Cook comparison -- it was very interesting.
It sounds like we won't be exploring Mars until we have a population of would-be explorers that is 1) worse off here than in space, 2) led by a captain with a penchant for the lash, and 3) drunk off their arse.
I think we already do have that population, even in western society. A lot of people live enough below the poverty line and with reduced life expectancies such that they'd happily gamble their lives on an adventure like this, irrespective of what some people might think about that attitude. It's still not an accessible activity for these people, though, unlike sailing ships probably were a few centuries ago. There are several reasons that come to mind, and the most obvious to me are:
Qualifications: Most astronauts at the moment (at least for agencies like NASA) need to be very qualified in one way or another. With few exceptions, people who are that qualified simply aren't below the poverty line. If space travel was at a point where the people running the ship didn't have to be rocket scientists (or similar), this might not be such an issue.
Social responsibility: Even if it's a myth in practice, most western societies are very tangled up in ideas of social responsibility at the moment. If you don't care about dying, it's irrelevant. The people who build the valve that fails and blows a hole in the side of your ship will still be prosecuted, or held liable. If not by you, it'll be your family, or your employer, or someone else, regardless of what you said before leaving. Everyone's been conditioned to be so paranoid about being blamed that the decisions of an individual in such a big operation become irrelevant.
Until these are sorted, I don't think there's much hope of knowingly risky spaceflight.
Having a "receipt" is pointless, except for extortion uses; it isn't a reliable indicator of the machine tabulation and can't be used for manual recounting.
Well, it's only pointless if it's returned to the voter. I think what was meant by the parent (which wasn't very clear) was to simply allow the voter to confirm that their correct vote is also recorded on paper. The receipt can then be automatically dropped into a paper ballot box. (Or alternatively if it wasn't correctly recorded on paper, the voter indicates this, the receipt and electronic vote are ignored, and the voter votes again.) If there's any significant controversy over the electronic counting, the collected paper receipts can be counted manually as a final authority.
People simply can't confirm that their vote has been correctly recorded inside a computer. I have a computer science degree, and that only makes me even more wary that it might not have been, perhaps because I'm aware of all the ways that it might be abused. There are too many levels of abstraction between the real world and computers, and it takes a lot of training and expertise to properly understand it. Most people will never be able to confidently understand it, either because they don't have the time, or they simply can't think in that way. But they can check that it's correct on paper, and then watch the paper be dropped into the ballot box.
It's pretty similar in many ways to the system you've described for Wisconsin, I think. What really matters is that both of these systems allow for manual recounts of voting papers that have been verified by the voters. Doing this returns the voting system back to a level where all voters can see and understand how it works on at least the most fundamental level where things will be decided in the case of any controversy. Understanding of the system by as many people as possible is where trust comes from, and trust in the election system is one of the big things that's ultimately needed for a fair and respectable democracy.
Apt and Yum seem to be the main software update mechanisms in use at the moment on Linux, but both seem to require you to download the entire application or library that you're updating.
My situation means that I'm stuck with a dialup connection for much of the time. I use Debian (etch/testing).
The most frustrating thing that I find with apt is that it's necessary to re-download the entire package list simply to find out what packages have been updated. If you like to subscribe to two sections (eg. I get both testing and unstable for a few packages that I want more recent), that's a good 45 minutes of downloading new package lists just to have the lists up-to-date... it's very offputting if I really only want to update a 200k package. And if you let them go stale for a few days before actually downloading/installing any updated packages, there's a more and more significant chance that some packages will have been updated further with the listed files having disappeared from the repository... requiring yet another apt-get update to find out what the filenames for the new versions are. (Often it's easier to just go to http://packages.debian.org/ and download them directly, but this defeats the purpose of actually having apt.)
A straightforward solution to this would be to support and provide diffs for package lists. Even if they were only left on the servers for a few days, it'd be easy enough to make sure I updated every few days. I note that this has actually been listed as a feature request since January 2002, but has unfortunately not been implemented (although it looks like there's now a patch being developed, which is promising). Using something like rsync (eg. via apt-rsync is also an option. It requires the mirror site to support rsync, though, which many don't. Also if lots of people suddenly used rsync to update package lists, I think those mirrors that do support it would stop very quickly, simply due to the CPU load that rsync puts on the server.
Diffs for actual package updates would be great, but I don't find the lack of them to be anywhere near as annoying as not having diffs for package lists, simply because I don't usually care about updating the majority of packages to be most up-to-date. Package lists, however, have to be updated frequently, irrespective of the size of the package actually wanted.
Just thinking about it now, it'd be nice if apt could more clearly indicate which packages need to be updated for an upstream release, versus those where the package maintainer has simply made a minor adjustment somewhere. eg. If I have openoffice.org 1.1.2-7 installed, then I really don't care about downloading 40-50MB to "upgrade" to openoffice.org 1.1.2-9, but I might care if it was going to 1.1.4-3, because that's a much more significant jump. That's something I can cope without for now, though.
I'm surprised you didn't list xfig, despite listing a port and a clone (and noting they were related to it).
I often use xfig to draw simple figures for latex documents that I write... I've always found the interface quite awkward to use, though..fig files are also a bit restricted, but conveniently they're often easy to edit by hand.
Evidently the discoverers have been withholding this information from the public since 2003 while they waited for full analysis.
Is there really a problem with not releasing immediately? The linked article is very biased to suggest there is, but it seems like quite a naive attitude to me.
They wanted to know more information first, and it's not exactly a piece of information critical to safety or people's future. It was always their risk that their discovery might have been overshadowed by someone else discovering it, but it was their risk to take.
It's common practice for researchers to hold off presenting something for many reasons, including wanting to have something extra to report or that they want to make sure that they understand what they have. It's also common for astronomers to make announcements that are prematurely mis-interpreted by the media... and right now there a thousands of news reports around the world speculating about the planet/asteroid argument all over again.
Be careful not to confuse idiocy with desperation. Employment law is there (among other reasons) to prevent employers from taking advantage of people who are desperate for work, and from making a hiring decision based on whoever consents to give up the most fundamental rights to the whim of the employer. You can't sign someone into lifetime slavery, either, even if they consent to it because you'll "protect their family" or some-such promise.
A lot of people will sign anything if they're desperate, especially if they don't think it'll get to a point where the employer can carry out such a threat. Weighing out a salary versus human rights appears to be a rational decision. It doesn't make it fair and it doesn't make it legal contract.
I can understand the Illegal Search by the Cop (as I've been made to under-go such a thing myself,) but as to the drug testing, unless you did not agree to such a thing when you started your job, well, it's kind of like having to deal with a Non-Compete clause. You agreed to it.
This isn't true. Contracts can contain illegal terms, and if they do then they're invalid. An obvious example would be if an employer inserted a clause saying they could kill you if your performance fell below a certain level. If an employee signed their life away like that for whatever reason, it's irrelevant. The contract clause is illegal, and any employer that followed through on it would be in a lot of trouble.
There are legal protections on the content of employment contracts to stop employers from demanding unreasonable conditions from their employees, current or future. It's also why we have things like minimum wage. Some rights can be given up in a contract, but others can't.
Whether a drug-testing clause is or isn't okay would depend on your local legislation. Some governments would definitely consider it a breach of personal rights, and would disallow an employer from deciding who to hire based on their acceptance of submitting to a drugs test. Chances are there would sometimes be exceptions with this, however. It might be acceptable, for instance, if it's an obvious safety issue on the job, and/or if there's reasonable cause for suspicion that you're taking drugs. An employer might have to provide convincing evidence for suspicion, however, regardless of what a contract states.
Personally I don't think that nearly enough is done to stamp out ridiculous and illegal clauses in contracts. This is exactly the same reason why we have hopelessly one-sided terms of service on shrink-wrapped software. There's very little, if any, penalty for putting in highly dubious or illegal clauses and then pressuring someone to agree to them.
It will make it harder to run telescopes, but also a number of navigational devices. The mention of the Glonass screwup is actually misleading - even if you abolish the leap second, you still have to have software in your satellites compensate for changes in Earth rotation rates - abolishing the leap second will not change that at all.
I haven't thought about this in detail, but I tend to agree with you. My impression is that abolishing the leap second probably won't do much to help any of the highly accurate systems that already have problems, because those systems will just have to reverse the way they're dealing with things.
Astronomy's a good example. Astronomers already have to cope with leap seconds (through concepts like Dynamical time) for calculating things like star and planetary positions in the past and future. Removing leap seconds would keep UT on track with where the planets are, but they'd simply have to correct for leap seconds elsewhere, keeping in mind that if we make time consistent, it doesn't make the Earth consistent. eg. Tasks like keeping Earth-based telescopes correctly aimed would need leap seconds to be subtracted instead of added.
Similarly, any complex system that's related to satellites, the Sun, or anything non-terrestrial, will still have to have a mechanism for converting.
The fact is that there are already at least two perceptions of time, and many more if you decide to look further. You'd hope that any systems that requires so much accuracy would already have been developed by people competent enough to clearly understand the concept of Time. Unless it can be shown that this will be astoundingly useful, I don't think it's worth doing. It'll create even more systems of time, which will lead to even more possible ways to make mistakes.
Sourceforge is littered with the remains of OSS projects that were fun to code and get working, but that nobody wants to maintain anymore.
What you've said about the administration problems for large projects is true, but I don't think it's necessarily a bad thing that there are lots of unfinished projects lying around places like Sourceforge.
A few months ago, I was looking for a library that would do something, but it just didn't exist. What I did find, though, was someone's Sourceforge effort from five years ago. It wasn't packaged very well, and it only covered about 70% of what I'd ideally want. I was able to contact the original author, and while he's still interested in it, he really doesn't have the time (or to some extent the expertise) to finish it.
Since then, I've decided to try to pick up where the previous developer left off. I've re-packaged the code, and now I'm thinking about extending it to cover what I wanted to do previously. I don't know how successful I'll be in finishing it off, and to be honest I think it's unlikely. But the fact that someone else made their own effort available, and occupying sourceforge, made it much easier for me to get my own effort underway.
Seriously, this looks like it was designed by an amateur software development team. This is meant to be the Firefox killer?
It's a beta. If they've been focusing on decent HTML and CSS support, being careful about security issues, and fixing all the other problems that IE has been plagued with, I don't really care about the interface at this point. I'd expect them to fix it before the final release, but it's silly to expect a final product if they haven't called it that.
Why is the menu Below the tabs. I find this inconsistent and confusing. Worst of all, there's no way to put it in it's proper position.
Something I've often wondered about Firefox (which IE7 seems to do as well from your screenshot) is why the URL bar is above the tabs. Each tab has a different URL, a different history, and a different page. To me at least, it'd make much more sense to have the URL bar (and back, forward, refresh buttons) placed below the tabs, to indicate how they're associated.
The UNIX mentality of small, modular programs doing one thing well can still be maintained while a graphical environment is running, but his criticism that "do everything kind of okay, interface with nobody" can't be taken as criticism: it's just the way that GUI stuff appears to the user*.
I don't think the GUI-ness of an interface has much to do with it. There are already several programming environments (often aimed at children) that are based around the organisation and connection of graphically-presented blocks, each of which does some function. Some of them are designed to give lots of graphical and visual feedback about what's happening, and hints about why something is or isn't working. Command Line Interfaces are a neat and efficient way to do similar things, but they're not the only way.
If there's a problem, I think it's that most people who are general users of computers don't want to have to learn or understand how to chain things like this together. They're after much more immediate results, and usually more intent on dealing directly with their problem rather than wanting to learn how to construct a program in detail. I can sympathise with that -- often if I buy something as a necessity, without being really interested in it, I'll settle for something that's recommended instead of taking the time and effort to get or build one suited exactly to me.
Also Google Groups features a quite confusing interface ever since the remake (and the remake was what added the non-usenet groups).
I found the new interface very confusing compared with the old one, but just today I was looking at it, and noticed the "Browse all of Usenet..." link in the lower right corner of the page.
That link points here, which is very similar to the old group browsing interface that they had, and one that I can use to find what I'm looking for much more easily.
I can kind of understand why they've been trying to change the interface, even though I don't like the new one as much. Anyone who's unfamiliar with Usenet will probably be much more familiar with the standard web catalogue format, where things are placed into nice, plain categories with clear titles, rather than systematic dot-separated hierarchical names. I prefer the second way, but a lot of beginners wouldn't have a clue what the difference is between alt., comp., rec., soc., and all the others.
I don't find this surprising. Lately, it seems that Yahoo has been getting some of the positive internet buzz that used to be reserved solely for Google.
They're definitely providing at least a couple of services which I'm surprised that Google isn't heavily involved in just yet.
One of them is YahooGroups, for running mailing lists (along with several additional group-like features latched on). I guess Yahoo picked up a lot of this market by default, especially after Listbot was shut down by Microsoft. The other is Yahoo Calendar, which I'm admittedly only just starting to play with, but I'm finding it useful.
The biggest reason that I'm surprised Google hasn't touched these areas is that they're both very search-oriented, or can be. Just about everything Google's done in the past has been based around some kind of searching, or generally helping people to find things. That's where Google's expertise is.
Slashdot Mirror
It sounds like the nature of nearly all software to me. There are a few projects, open source and closed source, that are of high quality, well and consistently maintained. The majority aren't, and I don't think their philosophy has much to do with it.
The big (long term) advantage I see with open source is that if the developers decide to ditch it, you always have the option of paying someone else to keep it maintained for you... or doing it yourself if you have the expertise. If closed source developers decide to kill their project, on the other hand, that option isn't at all certain. There are plenty of closed source software projects out there that are no longer maintained.
I think your problem is at least as much with the editors as with the journalists. I realise what you're saying, but journalists aren't all like that. I have a good friend who's an amateur astronomer, and she's also a freelance science writer. She has things published in daily newspapers and/or weekly magazines about every week or two. She gets paid a commission, but definitely doen't make enough to cover the time she spends on it. Her primary motivation for doing it is to simply try and get some science writing in the media at all.
The irony is that even though she knows exactly what she's talking about when writing something, and she puts a huge amount of effort into being careful and accurate about what she writes, she still has to fight with editors. Probably most of the articles she writes end up getting lazily re-written in some way, even if it's just re-typed for some reason with half the words messed up. (eg. In an article last week, they changed "mass spectrometer" to "massive spectrometer", which is completely different!) It's not uncommon for chunks of writing to be cut without proper consultation, and with no respect for the context, or how they might be changing the message of the article. On a couple of occasions, they've held back time-critical articles and published them weeks after they were actually relevant, as if they didn't even bother to check the content properly.
It's not just the editors in this case, either. It's the whole system that involves deadlines and priorities that the people in the business give themselves. Editors of regular media publications just don't get to be editors by knowing about or having much respect for science. In the cases I've mentioned above, they've made a broad descision to step up and help to publicise science, but in reality they don't really care too much about the specifics of what they're doing -- it's for show as much as anything.
Some journalists "exist to be published", but I think the main ones in that frame are the staff journalists who are being paid a salary. Those people are only a subset of all the writing that you're likely to see in many publications. The problems are likely to come especially at times when they're being asked to write about things they don't find particularly interesting... and there simply aren't many staff journalists out there who find scientific topics anywhere near as interesting as things like politics, crime or business, for instance.
Did you consider using Mono for hosting of .NET on a non-Windows platform? From the tone of your post, I'm guessing that your boss would have blocked it, but I'm just curious. I'm only just starting to look at Mono (just for fun right now), and I'm wondering how reliable it is compared with Microsoft's implementation.
Because the way Microsoft's going, they don't see it as a Word Processing document any more. They see it as an access point for exchanging information, and there's really no reason (in Microsoft's view) why there should be boundaries on what types of information or interaction should be found in what places.
I don't particularly like the way that Microsoft operates, and I have no doubt that their criticism of the Massachusetts decision is plain and simple FUD. As you say, it probably makes much more sense for a government entity, and probably most other entities, to use well defined, straight-forward formats for storing its important information. I think it's a bit short-sighted, though, to criticise Microsoft too much for mixing together VOIP and Word Processing.
They're just working with a different model from the traditional metaphor that most people are familiar with. (No, I don't think they're doing a very good job being clear about what metaphor, if any, they are using.) Really though, the only reason people presently think it's strange to mix Word Processing and Voice Over IP is because we've so far been conditioned to think of a Word Processing document as a printable piece of paper. There are plenty of other things with writing on them that aren't paper, and it's not necessarily true that a word processor shouldn't mimic or act like them just as much, if a user wants to use it that way.
In Microsoft's world, someone will be able to click an icon (or access a "thing", or whatever) without having to care about what it is. They'll simply get whatever it contains, whether it's printable typewriter-style text, or some kind of portal for speaking to someone else. Similarly, people will be able to construct whatever media they want, without necessarily needing to think about whether they're dealing with a word processor, a spreadsheet, or a video camera. The biggest barrier is people's ability to adapt from how they relate to things in the real world. Metaphors are really only needed for people who have prior experience with something else, and there are some valid arguments that metaphors cause more problems than they solve.
I don't know if this is an ideal way to go or not, but I think you really need to recognise that Microsoft is not necessarily limiting themselves to the traditional ways of using an Office Suite any more. For better or worse, this is one of the differences between Microsoft and Open Office.
The Feds aren't designed to move fast, but they really should have thought ahead so that they'd never have to.
I'm not a citizen of the USA. If I was, though, I wouldn't think it too unreasonable for the Federal Government to at least coordinate with the state governments, make sure they're prepared, and to help arrange preparations if necessary. This might easily involve helping to provide caches of suitable resources in strategic places (whether in the state or nearby), providing expertise to make sure that the response plans are feasible, coordinating agreements with neighbouring states, etc. It might also involve issuing warnings directly to citizens if the feds believe that state officials aren't doing their job properly, or are ignoring federal advice.
If the Federal Government had had a proper programme ages ago to ensure that state governments had identified, were aware of, and properly prepared for possible future disasters in their region (such as this one), it might never have been such a serious problem.
For what it's worth, Microsoft is (also) a New Zealand company. It was registered at the NZ Companies Office on 29 May 1991. It really is an American company, but I guess they at least have a claim to comment on New Zealand law.
I think one of the problems with this method at the moment is that most organisations don't provide clear information about exactly what someone can expect by checking the box. Often it's buried somewhere deep within the privacy policy, but it's not exactly obvious.
Before I check such a box, I like to be confident that I understand basic things like:
Most boxes don't actually do this. They just say inane things like "Click here to receive great deals from us and our partners in your email." I'd rather they said something like:
I guess it's a bit more verbose, but to me it's a hell of a lot clearer and more trustworthy. Then again, I realise that most people don't seem to think/care about this type of thing as much as I do. I'm sure I'm not the main target of many marketing people... I just get annoyed as collateral damage.
I remember the '90's. I remember that it was actually legal for people and businesses to lend or rent, or even re-sell the software they'd purchased to other people. The only condition was that it wasn't allowed to be run in more than one place at a time. Locally, we even had rent-by-mail companies that would take out full page advertisements in magazines, and post you software to use for a limited amount of time before you were (legally) required to uninstall it and return it.
Software companies -- not even Microsoft, but especially Microsoft -- went to great lengths to inform their customers that they were allowed to install Microsoft Word from their work onto their home PC, as long as it was only being used on one PC at a time. After all, it was clear that the software was licenced for use to the person who'd paid for it, so they were allowed to use it wherever it was most convenient for them without having to pay multiple times.
Some software had basic copy protection (eg. flight simulator games that asked you to quote a random word from page 215 of the manual), but there was rarely serious DRM. Software companies were quite clearly concerned about software piracy, but they weren't trying to wipe out customer's existing rights to solve that problem. On the contrary, most software companies and consumers actually acted as if they "trusted" each other, compared with today.
Somehow, this whole attitude has been lost in recent years, at least in commercial software. A major part of it seems to have been about when the media barons leapt to digital and brought all their annoying views on draconian copy protection with them. There's no way in hell you'd be allowed to casually install a work copy of MS Word on a home PC today, unless your company had a special agreement that they'd paid extra for. Even with this, there would still be DRM loopholes to jump through. You'd probably end up in jail for 15 years as an example to other "criminals" if the Microsoft-sponsored BSA lawyers had their way.
Times change, I guess.
I coincidentally watched Dark City on video a couple of weeks before The Matrix was released. I don't remember either of them very well any more, but I do remember thinking, as I was watching it, that the Matrix seemed like a rip-off of Dark City in many ways. It was probably just coincidence in hindsight.
Of course, as others have also pointed out, Dark City itself was preceeded by Metropolis.
I think the article sums up this area quite nicely:
I'm glad the author included this note. It indicates quite clearly why a lot of companies see it as being so important to be cautious of lawsuits. It's not the possibility of losing that matters, it's the cost of fighting. If the cost of fighting is more than a company can afford, it just make sense to settle, and then the overly litigous company wins. Even enlightened CEO's often want indemnification--not because they think a company like SCO has a hope of winning a lawsuit, but because they're concerned about how much it might cost if they're even targeted.
Personally I think there need to be some changes to the system so that those who abuse it in this way are penalised much more, and also more quickly. The fact that SCO was both able to be so noticed in the first place, and can even continue dragging its corpse around today to threaten people with its stench without the likelihood of serious charges being brought against it, its executives or its legal team, suggests that not quite enough measures yet exist.
As much as I agree with the general sentiment of this thread about marketing (which I frequently get sick of), there are times where I actually appreciate it. I don't like marketing and advertising that's in my face, and I don't like marketing that lies to me. But some marketing material is just out there to be informative for people who want it and ask for it, without being in anyone's face.
Better results for me means being fully informed about all the relevant options I have, at a time and level of detail of my choosing. This is also a type of marketing, and it's one that I respect. I do know some marketing people who focus on this goal, and I appreciate it.
I actually like the way that the shoe salesman walks up to sell me a shoe when I walk in. I really have no idea what I want and it's not a decision I want to make. What I care about is trusting the guy to sell me something that works, and that's what will get me to come back again and again. That's also marketing. The guy's job relies on him selling lots of shoes, but he knows that his best approach is just to be honest with people. (and to chat, and joke, and so on.)
I also quite like the way that Amazon suggests books for me to read. It's only there when I ask for it, it often offers good suggestions, and every so often they might get a sale out of it. That's the type of marketing that I like.
Thanks for the Cook comparison -- it was very interesting.
I think we already do have that population, even in western society. A lot of people live enough below the poverty line and with reduced life expectancies such that they'd happily gamble their lives on an adventure like this, irrespective of what some people might think about that attitude. It's still not an accessible activity for these people, though, unlike sailing ships probably were a few centuries ago. There are several reasons that come to mind, and the most obvious to me are:
Until these are sorted, I don't think there's much hope of knowingly risky spaceflight.
Well, it's only pointless if it's returned to the voter. I think what was meant by the parent (which wasn't very clear) was to simply allow the voter to confirm that their correct vote is also recorded on paper. The receipt can then be automatically dropped into a paper ballot box. (Or alternatively if it wasn't correctly recorded on paper, the voter indicates this, the receipt and electronic vote are ignored, and the voter votes again.) If there's any significant controversy over the electronic counting, the collected paper receipts can be counted manually as a final authority.
People simply can't confirm that their vote has been correctly recorded inside a computer. I have a computer science degree, and that only makes me even more wary that it might not have been, perhaps because I'm aware of all the ways that it might be abused. There are too many levels of abstraction between the real world and computers, and it takes a lot of training and expertise to properly understand it. Most people will never be able to confidently understand it, either because they don't have the time, or they simply can't think in that way. But they can check that it's correct on paper, and then watch the paper be dropped into the ballot box.
It's pretty similar in many ways to the system you've described for Wisconsin, I think. What really matters is that both of these systems allow for manual recounts of voting papers that have been verified by the voters. Doing this returns the voting system back to a level where all voters can see and understand how it works on at least the most fundamental level where things will be decided in the case of any controversy. Understanding of the system by as many people as possible is where trust comes from, and trust in the election system is one of the big things that's ultimately needed for a fair and respectable democracy.
My situation means that I'm stuck with a dialup connection for much of the time. I use Debian (etch/testing).
The most frustrating thing that I find with apt is that it's necessary to re-download the entire package list simply to find out what packages have been updated. If you like to subscribe to two sections (eg. I get both testing and unstable for a few packages that I want more recent), that's a good 45 minutes of downloading new package lists just to have the lists up-to-date... it's very offputting if I really only want to update a 200k package. And if you let them go stale for a few days before actually downloading/installing any updated packages, there's a more and more significant chance that some packages will have been updated further with the listed files having disappeared from the repository... requiring yet another apt-get update to find out what the filenames for the new versions are. (Often it's easier to just go to http://packages.debian.org/ and download them directly, but this defeats the purpose of actually having apt.)
A straightforward solution to this would be to support and provide diffs for package lists. Even if they were only left on the servers for a few days, it'd be easy enough to make sure I updated every few days. I note that this has actually been listed as a feature request since January 2002 , but has unfortunately not been implemented (although it looks like there's now a patch being developed, which is promising). Using something like rsync (eg. via apt-rsync is also an option. It requires the mirror site to support rsync, though, which many don't. Also if lots of people suddenly used rsync to update package lists, I think those mirrors that do support it would stop very quickly, simply due to the CPU load that rsync puts on the server.
Diffs for actual package updates would be great, but I don't find the lack of them to be anywhere near as annoying as not having diffs for package lists, simply because I don't usually care about updating the majority of packages to be most up-to-date. Package lists, however, have to be updated frequently, irrespective of the size of the package actually wanted.
Just thinking about it now, it'd be nice if apt could more clearly indicate which packages need to be updated for an upstream release, versus those where the package maintainer has simply made a minor adjustment somewhere. eg. If I have openoffice.org 1.1.2-7 installed, then I really don't care about downloading 40-50MB to "upgrade" to openoffice.org 1.1.2-9, but I might care if it was going to 1.1.4-3, because that's a much more significant jump. That's something I can cope without for now, though.
I'm surprised you didn't list xfig, despite listing a port and a clone (and noting they were related to it).
I often use xfig to draw simple figures for latex documents that I write... I've always found the interface quite awkward to use, though. .fig files are also a bit restricted, but conveniently they're often easy to edit by hand.
Is there really a problem with not releasing immediately? The linked article is very biased to suggest there is, but it seems like quite a naive attitude to me.
They wanted to know more information first, and it's not exactly a piece of information critical to safety or people's future. It was always their risk that their discovery might have been overshadowed by someone else discovering it, but it was their risk to take.
It's common practice for researchers to hold off presenting something for many reasons, including wanting to have something extra to report or that they want to make sure that they understand what they have. It's also common for astronomers to make announcements that are prematurely mis-interpreted by the media... and right now there a thousands of news reports around the world speculating about the planet/asteroid argument all over again.
Be careful not to confuse idiocy with desperation. Employment law is there (among other reasons) to prevent employers from taking advantage of people who are desperate for work, and from making a hiring decision based on whoever consents to give up the most fundamental rights to the whim of the employer. You can't sign someone into lifetime slavery, either, even if they consent to it because you'll "protect their family" or some-such promise.
A lot of people will sign anything if they're desperate, especially if they don't think it'll get to a point where the employer can carry out such a threat. Weighing out a salary versus human rights appears to be a rational decision. It doesn't make it fair and it doesn't make it legal contract.
This isn't true. Contracts can contain illegal terms, and if they do then they're invalid. An obvious example would be if an employer inserted a clause saying they could kill you if your performance fell below a certain level. If an employee signed their life away like that for whatever reason, it's irrelevant. The contract clause is illegal, and any employer that followed through on it would be in a lot of trouble.
There are legal protections on the content of employment contracts to stop employers from demanding unreasonable conditions from their employees, current or future. It's also why we have things like minimum wage. Some rights can be given up in a contract, but others can't.
Whether a drug-testing clause is or isn't okay would depend on your local legislation. Some governments would definitely consider it a breach of personal rights, and would disallow an employer from deciding who to hire based on their acceptance of submitting to a drugs test. Chances are there would sometimes be exceptions with this, however. It might be acceptable, for instance, if it's an obvious safety issue on the job, and/or if there's reasonable cause for suspicion that you're taking drugs. An employer might have to provide convincing evidence for suspicion, however, regardless of what a contract states.
Personally I don't think that nearly enough is done to stamp out ridiculous and illegal clauses in contracts. This is exactly the same reason why we have hopelessly one-sided terms of service on shrink-wrapped software. There's very little, if any, penalty for putting in highly dubious or illegal clauses and then pressuring someone to agree to them.
I haven't thought about this in detail, but I tend to agree with you. My impression is that abolishing the leap second probably won't do much to help any of the highly accurate systems that already have problems, because those systems will just have to reverse the way they're dealing with things.
Astronomy's a good example. Astronomers already have to cope with leap seconds (through concepts like Dynamical time) for calculating things like star and planetary positions in the past and future. Removing leap seconds would keep UT on track with where the planets are, but they'd simply have to correct for leap seconds elsewhere, keeping in mind that if we make time consistent, it doesn't make the Earth consistent. eg. Tasks like keeping Earth-based telescopes correctly aimed would need leap seconds to be subtracted instead of added.
Similarly, any complex system that's related to satellites, the Sun, or anything non-terrestrial, will still have to have a mechanism for converting.
The fact is that there are already at least two perceptions of time, and many more if you decide to look further. You'd hope that any systems that requires so much accuracy would already have been developed by people competent enough to clearly understand the concept of Time. Unless it can be shown that this will be astoundingly useful, I don't think it's worth doing. It'll create even more systems of time, which will lead to even more possible ways to make mistakes.
What you've said about the administration problems for large projects is true, but I don't think it's necessarily a bad thing that there are lots of unfinished projects lying around places like Sourceforge.
A few months ago, I was looking for a library that would do something, but it just didn't exist. What I did find, though, was someone's Sourceforge effort from five years ago. It wasn't packaged very well, and it only covered about 70% of what I'd ideally want. I was able to contact the original author, and while he's still interested in it, he really doesn't have the time (or to some extent the expertise) to finish it.
Since then, I've decided to try to pick up where the previous developer left off. I've re-packaged the code, and now I'm thinking about extending it to cover what I wanted to do previously. I don't know how successful I'll be in finishing it off, and to be honest I think it's unlikely. But the fact that someone else made their own effort available, and occupying sourceforge, made it much easier for me to get my own effort underway.
It's a beta. If they've been focusing on decent HTML and CSS support, being careful about security issues, and fixing all the other problems that IE has been plagued with, I don't really care about the interface at this point. I'd expect them to fix it before the final release, but it's silly to expect a final product if they haven't called it that.
Something I've often wondered about Firefox (which IE7 seems to do as well from your screenshot) is why the URL bar is above the tabs. Each tab has a different URL, a different history, and a different page. To me at least, it'd make much more sense to have the URL bar (and back, forward, refresh buttons) placed below the tabs, to indicate how they're associated.
I don't think the GUI-ness of an interface has much to do with it. There are already several programming environments (often aimed at children) that are based around the organisation and connection of graphically-presented blocks, each of which does some function. Some of them are designed to give lots of graphical and visual feedback about what's happening, and hints about why something is or isn't working. Command Line Interfaces are a neat and efficient way to do similar things, but they're not the only way.
If there's a problem, I think it's that most people who are general users of computers don't want to have to learn or understand how to chain things like this together. They're after much more immediate results, and usually more intent on dealing directly with their problem rather than wanting to learn how to construct a program in detail. I can sympathise with that -- often if I buy something as a necessity, without being really interested in it, I'll settle for something that's recommended instead of taking the time and effort to get or build one suited exactly to me.
I found the new interface very confusing compared with the old one, but just today I was looking at it, and noticed the "Browse all of Usenet..." link in the lower right corner of the page.
That link points here, which is very similar to the old group browsing interface that they had, and one that I can use to find what I'm looking for much more easily.
I can kind of understand why they've been trying to change the interface, even though I don't like the new one as much. Anyone who's unfamiliar with Usenet will probably be much more familiar with the standard web catalogue format, where things are placed into nice, plain categories with clear titles, rather than systematic dot-separated hierarchical names. I prefer the second way, but a lot of beginners wouldn't have a clue what the difference is between alt., comp., rec., soc., and all the others.
Good point. Until checking just now, I hadn't realised that they'd extended it to anything more than a wrapper around Usenet.
They're definitely providing at least a couple of services which I'm surprised that Google isn't heavily involved in just yet.
One of them is YahooGroups, for running mailing lists (along with several additional group-like features latched on). I guess Yahoo picked up a lot of this market by default, especially after Listbot was shut down by Microsoft. The other is Yahoo Calendar, which I'm admittedly only just starting to play with, but I'm finding it useful.
The biggest reason that I'm surprised Google hasn't touched these areas is that they're both very search-oriented, or can be. Just about everything Google's done in the past has been based around some kind of searching, or generally helping people to find things. That's where Google's expertise is.