I have mixed feelings about Tornquist. He created The Longest Journey which was absolutely amazing, particularly for its story. My favorite game ever. Vivid, detailed characterization, intricate world-building, compelling plot. The tech wasn't impressive (3D figures superimposed on 2D backdrops), but the story was so great that I didn't care.
Then came the sequel, Dreamfall. Oh. My. God. The game was a lot prettier, a good deal more tech glitz. But the UI was atrocious (horrible camera control, unplayable on PC without a USB controller), gratuitous fighting scenes built in (complete with lousy combat controls), and the puzzles (such as they were) didn't make sense. Worse, the plot was incoherent at many crucial points, and the main character (Chloe) completely failed to engage my sympathy or even interest. I got to the end and was sorry she hadn't died permanently somewhere along the way.
Dreamfall had the most severe case of sequel-itis I've ever seen. The original was amazing, astounding, wonderful, and sold a bazillion copies. Then the corporate types took over and threw a ton of cash at the sequel, and it sucked hard. The only comparable thing I can think of? Indiana Jones -- Raiders of the Lost Ark was terrific, and Temple of Doom sucked so hard that nobody ever plays it on TV, not even at 4 AM to fill up time. That's how Dreamfall was.
I have hopes for the third TLJ installment -- after all, The Last Crusade rescued Indiana Jones from one-hit wonder status. It could happen again. But then I think of how the Matrix series went downhill, and fear.
Under blanket licensing, how do I reward artists with good music preferentially to those who suck?
We could always set up the system to reward artists in proportion to 1) the number of times their songs are downloaded, and 2) the average rating of the song.
So:
Step 1: Assess blanket fee (ie tax)
Step 2: Establish a way to track the number of times a song is downloaded and/or played. This could be either a central repository tracking downloads or a distributed system that reports "+1 play Nobody Cares by Levinhurst" to a central server. Obviously lots of potential for abuse if the central server isn't secured properly.
Step 3: Track user ratings for each song. As a simplistic example, scale of 1-5 stars.
Step 4: Work out a formula based on that data, e.g. (base rate * average rating) * total number of plays, calculated at intervals (e.g. monthly).
Step 5: Profit! For the artists, that is.
So if, for example, the base rate is 0.02 (two tenths of a cent) and a single song gets an average rating of 2.3 and plays 201,435 times in April, then sometime in early May the artist gets a payment of (0.02 cents * 2.3 rating) * 201,435 plays = $9,266.01. The minimum rating (e.g. for a song that hasn't been rated by anyone yet) would be 1, to leave the base rate untouched. An artist with a really popular song, highly rated song would get more cash, for example (0.02 cents * 5) * 1,000,437 plays $100,043.70, cha ching! Whereas an artist with a lemon would get fairly little (0.02 cents * 1 rating) * 57 plays = $1.14, doh.
Obviously there are all kinds of details that would have to be worked out. Would the base rate be indexed to inflation? Who would be responsible for allocating the funds? What measures could be taken to prevent crappy artists from using bots to artificially inflate their download/rating? What would an appropriate base rate be? What could we do to account for cost-of-living differences between cheap places and expensive places?
And so on. But don't assume that blanket licensing means everyone gets paid exactly the same. There are lots of ways it could work, and dividing the pot equally amongst all comers is pretty much the dumbest approach.
Professional apps (CAD/simulators/visualizations...) make up the majority of the OpenGL market and they have to be supported for decades (no, military or airlines do not buy a new training system every two years...)
So breaking compatibility is deal breaker.
Hmm. If OpenGL is being maintained primarily for CAD developers and other people who require stable, long-term APIs, then perhaps the people who maintain the API should explicitly announce that as their goal.
I'm not trying to be snarky. If the needs of game developers are so incompatible with the needs of CAD developers, why not declare OpenGL for CAD and start a new group to focus on the needs of game developers?
There'd probably be a lot of resistance to that, though. Not to mention the many difficulties involved in assembling a group and producing a spec that the developer community (and their employers) can agree on. But it might be worth it.
[1] A fragment is a fancy name for a voxel defined in clip space. After shading and occlusion, the remaining fragments become rasterised as pixels. Thus, the term 'pixel shader' is rather inaccurate.
O_O
I suppose I shouldn't be surprised when people start speaking in alien languages... I mean, it's Slashdot!
Or perhaps you can enlighten me with a use case for a hobbyist website that requires SSL.
Hi! I have a domain which I acquired solely for the purpose of supplying myself and my family with webmail. I didn't want to have to put up with crappy advertisements, even the politely unintrusive ones in Gmail.
I'd really like to encrypt the log-in sessions for my webmail client. I understand that the email itself is not going to be encrypted once it's sent out. But as it stands, my EMAIL PASSWORD is being transmitted in clear text every time I log in. That sucks. Anyone on any server between me and my web host can get my email password, and they don't even have to go to the trouble of setting up a Man in the Middle attack, because it's completely 100% insecure.
In order to rectify this, I would need to:
Pay $2.50 extra per month for a unique IP address ($30 per year)
Pay $45 per year for an SSL certificate signed by RapidSSL, the hosting company's preferred SSL CA. They don't accept certificates from anyone else, and I can't install one manually, because I don't have shell access. It's a shared hosting account, which is all I can afford.
Install my own webmail client, because I can't install a certificate for the one they provide, which runs on a different server. Self-installed software is not supported, so I would essentially be giving up the technical support I've paid for.
All that just to secure my email password? Screw that. I'll keep doing it insecurely and hope to hell nobody hijacks my email to send out spam.
... virtually immediate obsolescence of any printed work... I'd like to point out that obsolescence varies by field. Sure, in the hard sciences (physics, chemistry, etc) research gets superseded very fast. But there are plenty of other fields out there, and not all of them work that way.
For example, I'm a medievalist. The people I study have all been dead for centuries, and genuinely new data are rare. Once every few years somebody will find a lost manuscript or something, but for the most part we're working with known, thoroughly studied information. Our research doesn't churn; it accretes. I routinely consult articles that are decades old, and in one instance I can think of, I actually cited an article that was over a hundred years old. New research is important too, but it tends to take the form of a new angle on existing data.
Other fields have their own tempos, I'm sure. It's a mistake to assume that all academic fields work alike.
ZDNet is a huge company, and I'm quite sure they've got a staff of full time web developers. It seems rather doubtful that the author of the article had ANY say whatsoever in the site design.
Which is a pity, because I bet he'd do a better job.
I took some measurements. First, some details on my setup. I'm using Firefox (2.0.0.14, because Firebug isn't working smoothly in 3.x yet). I have my browser maximized. I'm using a monitor with a resolution of 1050x1680 (that's a widescreen monitor rotated into portrait orientation so I can read more of a web page at once).
Under those conditions, the web page comes out at 1034 pixels wide by 2985 pixels tall (3,086,490 pixels total). The actual content of the article, which I measured using the Web Developer Toolbar's ruler function, fits into an area of 450 by 860 pixels (387,000). From which we can work out that only 12.5% of the page's area is being used for the actual article. The remaining 87.5% of it is given over to navigation, branding, and advertisements.
Navigation is important, branding is okay, and advertisements are tolerable in small amounts. But in this case, they've totally overwhelmed the actual content of the page, which is stupid. An article of this length should be split into no more than 3 pages, tops. 2 pages would be more reasonable, and it could be done in one page without undue strain, especially since a lot of the article consists of screenshots which don't need to be read closely.
If I read ZDNet on a more regular basis, I'd probably fiddle with their design. A judicious application of "display: none !important" and some other CSS in userContent.css would clean it right up; the at-rule @-moz-document domain(example.com) lets you target a particular site and apply your own CSS rules to their design any time you visit. The !important keyword makes your own rules more important than theirs. I don't do it for many sites, because it takes time and effort. But for the ones I care enough to read often (Salon.com, and, ahem, Slashdot) it's worth it.
The "ordinary" rows have a background color of pure white. The "striped" rows have a background color of #F5F5F5, a very light grey. I'm all in favor of subtle design, but there is such a thing as being too subtle.
Perhaps the stripes did not help noticeably because there was insufficient contrast between the rows?
Parent wrote:
Also, the one about finding MP3s on the Windows partition is not that easy - you could simply copy the files across with the Ubuntu migration assistant, but what if they're in a non-standard place? Indexing the Windows filesystem to quickly find these might help, but building the index could take some time. However, it would probably be enough if there was some feature in Ubuntu that scanned for existing partitions and said (based on partition type and a few key directories/files) that 'this looks like a Windows partition, it's available on the desktop through this icon', and ideally did a special symbolic link for the My Documents or similar (though that's tough as it's per-user under Windows - which user should this use). That solution to her problem is WAY more complicated than it needs to be. Here is the description of the problem she ran into, from the article:
Erin looked in her Music folder (created by default in Ubuntu), and her home folder and the desktop. She avoided the "492.8GB Media", which is my Windows partition, and the many strange and unhelpfully-named folders in "Filesystem" could only have scared her. She minimised Brasero and went to the "Places" menu for the first time, launching the search program. She tried searching for music in her home directory and music folder, but found nothing, and didnâ(TM)t try "Filesystem" or "492.8GB Media". This looks to me like a labeling problem. She knew it was in Windows, but the labels didn't identify the partitions clearly. A more helpful label would have been "Windows Disk" or similar. Since the vast majority of Windows installations are on either FAT32 or NTFS partitions, and it's exceptionally uncommon to find any other kind of OS using those partition types, it would be very easy to make the Ubuntu installer use a label like "Windows Disk" instead of just saying how big the disk is. To be extra certain that the partition really is for Windows, it could check for the presence of a Windows boot record in the MBR on the drive. This would only have to be done once (when the label is chosen during installation), and it'd be helpful.
In the same way, the label "Filesystem" isn't descriptive to a non-techie. She found the search interface, but didn't recognize that label as the one for searching everywhere. How about "Entire Hard Drive" or "All Hard Drives" or, as the article itself suggests, "Whole Computer" instead?
Indexing the partitions, migrating data, all that is potentially useful too, and may well be worth pursuing. But it's still a lot more complex than just picking some friendlier, more descriptive labels and adding a teeny bit more logic to the installer to apply them correctly.
I was very impressed with his work on Pan's Labyrinth, too.
I do have one reservation, though. Del Toro is primarily known as a director of horror films. The vast majority of his work is pretty seriously dark and violent. There are definitely some dark moments and some scary/violent scenes in The Hobbit (such as: the troll attack, riddles in the dark with Gollum, spiders in Mirkwood, and of course the Battle of Five Armies). But there are also a lot of light, delightful scenes (such as: songs in Rivendell, lunch with Beorn, seeing butterflies above Mirkwood, the kindly reception at Lake Town, and so on).
I may be going out on a limb here, but the overall tone of the book slants more towards "delightful" than "scary". Del Toro has amply demonstrated that he can do "scary". But can he do "delightful" just as well? If he can, we're in for a treat. If not, well, who knows what it'll be like? I'll definitely be interested to see what he comes up with; I just hope he does justice to the pleasant stuff as much as the unpleasant stuff.
Just yesterday Marketplace conducted an interview with the Craigslist CEO. You can listen to the interview on their site (it's the "listen to story" link, not the "listen to show" link). It's a pretty interesting interview -- one of the questions was "why aren't you running advertising?" And the answer was "because it's annoying and we think annoying our customers is stupid." When I heard him say that, I sat back and thought "Yeah! Nice to hear a CEO who really gets it."
They're a profitable company; but not madly, hugely, enormously profitable the way they could be. I like that. Capitalism offers powerful incentives to participate in the economy, which is generally good; but it also has a tendency to encourage empire-building and delusions of grandeur. So it's nice to see a company once in a while which isn't hell bent on world domination.
I wish their pages were a bit better designed, though. I appreciate the focus on navigation, but would it kill them to put borders around some of the boxes full of lists, to visually group them? And they could make some improvements for screen reader users, especially adding headings so that blind visitors could have their screen readers jump to the headings in the page rather than having to wade through everything that comes before the part they're interested in.
... beginning college students typically don't know what constitutes "good research". And they tend to be very trusting, not just of Wikipedia, but of anything on the Internet.
A few years ago I had a student turn in a paper arguing that the drinking age should be lowered to 18. One of the claims the paper raised was that drinking ages are lower in many European countries, and that they have healthier drinking cultures. That's probably true, but the source that the student cited to back up the point was totally inadequate. It was a two paragraph account of German drinking habits. The account was based on an interview with an unnamed exchange student. It was written down by an anonymous high school student. And it was put up on the web as a really badly designed web page. Let's see - anonymous author, anonymous interview subject, obviously done as part of a high school assignment, very short, no details, and badly presented. Not exactly the world's most credible source. I made the student go find a more thorough account of European drinking habits written by an identifiable human being and vetted by some kind of editor.
That's a fairly typical example. However, I don't think it's anything worth getting upset about. Students have long been overly credulous. Heck, people in general are overly credulous. It's always been possible to go out, find crappy information, and blindly accept it. Wikipedia (and more broadly the Internet) just make that easier. Yes, there's a lot of GOOD info out there on the web, too, but finding it can be very difficult.
That being the case, I try to integrate assignments about how you do research, and what constitutes a good source, what Internet sources are good for, and when you might want to hit the library and dig a little deeper. It's really a necessity. The students don't know how to do research; therefore, we need to teach them. Many schools are beginning to recognize this -- over the last ten years or so the number of positions at academic libraries for "instructional librarians" has skyrocketed. They visit other teachers' classes and teach lessons on search techniques, evaluation of sources, give tours of the specialized databases the university subscribes to, and so on. Some schools are even beginning to offer complete courses on information literacy. I think we'll probably see a good bit more of this over the next few years.
George Lucas' contract with Fox gives him the right to make money off the merchandise.
Andrew Ainsworth's contract does not; he could have requested a share, but he chose not to. His own dumb fault. There's a more informative article over at the Telegraph, which has two important bits of info.
1) There was no formal contract. In the absence of a formal assignment, it is unclear who owned the rights to the design of storm trooper armour. You could argue it either way.
2) Ainsworth is a British citizen who lives in Britain, and always has done. Under UK law, the "design right" in the shape of a physical object (which is NOT the same as copyright) expires after just fifteen years. So, in Britain, he's clear to make and sell just as much of the armour as he likes. It's possible that he may not be able to call it "Storm Trooper" armour; Lucas may have a trademark on that name. But he can definitely call it something more generic, e.g. "space trooper armour" or "space armour" or "sci-fi armour" or something.
Note that this is design right, not copyright. UK law recognizes a difference between the two. Design right applies to the shapes of things.
I'm interested in Mr. Ainsworth's defense. According to the article, he has two claims in his defense. First, "... that the intellectual property rights to the designs have expired." The Copyright, Designs, and Patents Act of 1988 specifies that design right (as distinct from copyright) lasts 15 years from when the design was first recorded or a finished product first made. Since storm troopers appeared in the very first Star Wars in 1977, and had to have been made in 1976 for the filming, I'd say that's a pretty good defense.
The article also says he claims that "if they [the rights] do still exist, as the designer he [Mr. Ainsworth] owns them rather than Mr Lucas." That seems rather more dubious to me. Mr. Ainsworth accepted a payment of 30,000 pounds for his work. Even in the absence of a formal contract, that would strongly suggest that he did it as a commission, in which case ownership of the rights would fall to the commissioner (Mr. Lucas in this case).
It'll be interesting to see how it comes out; and for once we won't have to wait too long, since the case is expected to take ten days.
Nuts. Posted before reading the whole thing - that's got some analysis highlights in it, but mostly it's a list of larger reports you can purchase from Gartner if you want.
The linked blog article is okay if you want a summary, but if you'd prefer, you can check out the complete document. Here's a PDF link to Gartner's full analysis: The State of Open Source 2008.
No skin off my back... I haven't actually paid attention to a commercial for years, and I only read print ads that are in scientific and tech related publications.
Interesting. I'm sure the marketers are pleased. The less conscious you are of their message, the less capable you are of resisting it.
Have you considered putting a table right next to the copier?
Alternatively, if there's one already there, have you put coasters on it, as a hint?
And if it's got coasters already, have you considered purchasing a cheap mug, drinking coffee out of it just once so it'll have an authentic ring-stain in the bottom, and then setting it on one of those coasters permanently as an added hint?
Failing that, have you taken a bunch of tennis balls, cut them in half, duct taped them to the top of the copier and spray painted them the same beige as the rest so there's no flat place to put drinks?
Further, have you considered sneaking into their cubicles by dead of night and supergluing their cups and mugs to the desk?
If all else fails, have you considered supergluing your coworkers themselves to their desks? I bet their productivity would go up. The smell might get bad after a while, though...
I've RTF preprint too, and I'm not sure this is a fair criticism. You've misunderstood something - specifically, you write:
They chose the brands that they chose because Apple was rated as more creative a brand than IBM in pre-test
Which is incorrect. The researchers chose the brand before the pre-test because they themselves believed it to be associated with creativity. The purpose of the pre-test was to verify that the researchers were correct in believing that the test participants perceive the Apple brand as associated with creativity. They also tested the participants' associations of the IBM logo (which was the other logo they showed) to see whether it too was associated with creativity, and found that it was not.
The end goal of the study was to measure whether or not exposure to a familiar brand would cause people to exhibit characteristics associated with that brand. Having verified that the Apple brand was associated with creativity in the population they were studying, and that the IBM logo was not, they primed their participants by flashing a logo at them for 80 milliseconds as part of a video, then administered a standard psych test measuring creativity (the Unusual Uses test - it was first put together in 1958 and has been thoroughly validated since then). There were 341 participants (190 male, 151 female). Out of those:
Half got the IBM logo, half got Apple.
All participants were asked to complete a meaningless task (crossing out E's in a text); half of the participants were asked to do so before the Unusual Uses test, half after. That part was to measure whether the effect of the logo, if any, dissipated rapidly.
They also asked participants after they had completed the test whether they had noticed any images in the video shown beforehand; except for the 80 millisecond flash of logo, it consisted of abstract testing patterns. Not one single participant reported being aware of any recognizable images in the video, so it really was subliminal.
Their analysis of the results showed statistically significant differences between those shown IBM and those shown the Apple logo. Specifically, the people who got Apple produced a greater number of responses on the unusual uses test, which is a quantitative difference that nicely demonstrates their desire to be creative if not the actual creativity. The researchers also had judges independently rate the responses on a qualitative scale. The judges didn't know which logo the participants they were evaluation had seen. The researchers then did statistical analysis on the qualitative judgments, which showed that the judges consistently rated the responses of the Apple participants as more creative than those of the IBM participants.
Then they did it again, two more times in fact, with different brands and traits.
They had lots of checks and counter-checks built into it, and they used some fairly sophisticated statistical analysis on the results, including some analysis of variance (ANOVA) checks. As is common with psychological studies, the test participants were all undergraduates taking an intro to psych class. So the population was not particularly diverse in terms of age. But on the whole, it looks to me like a pretty well designed study.
Yes, that's true. Human judgment about what constitutes high quality information helps a lot.
Regarding self-published material - yes, it's rare in libraries. I would add though that the fact that something was self-published doesn't necessarily mean that it's low quality. It could just mean that it's not commercially viable. Example: genealogy. If I write a genealogical history of families sharing the last name "Potter" who settled in Chillicothe, Virginia, during the last 150 years... that may well be very interesting and useful to local historians and genealogists from Chillicothe and surrounding towns. But it's not really going to have a large enough market for a commercial publisher to be interested. So instead you print up, say, 30 copies with a vanity press, and distribute them to local libraries and churches.
Hi, I'm a librarian.[1] I appreciate your response, and I'm glad you find us useful. I'd just like to elaborate on one of your points. You wrote "I like going to the library just to browse and to see what I can find." (Emphasis added). This is one point where physical libraries still have a distinct advantage over the Internet.
People who are trying to retrieve information have three basic types of queries.
They know exactly what they're looking for. (A "known item" search, e.g. "I want a transcript of the Obama-Clinton debate from Austin last week", which results in one document).
They know roughly what they're looking for. (A "known class" search, e.g. "I want to read essays on Kierkegaard's philosophy", which results in a reasonably well-defined group of documents).
They don't have a clear idea what they're looking for. (Called "browsing," e.g., "I'd like to learn about world history.", which results in a vastly huge set of documents that might potentially meet the need).
The Internet is pretty good for known-item searches. Especially if the item has indexable text in it. Other types of information are harder. Quick! Using Google Images, find me a picture of a sheep facing left at sunset.
The Internet is less good at delivering focused results for a known class search. It can retrieve relevant documents, but there's a good chance that it will also retrieve lots of unrelated or only tangentially related things. Which means you have to spend ages sorting through a giant list of search results to find what you really want. Specialized databases tend to produce much more focused results, of course, but most of those aren't freely available.
And lastly, the Internet is lousy for browsing. Browsing is about finding out what's available within a very broad class of stuff. Search engines can tell you that documents share keywords; they can't tell you for certain that the documents are actually about similar things. And within the search results, they're organized according to (roughly) how popular they are, as measured by how many sites link to them. They're not organized based on their similarities to or differences from one another. Compare to a library, where you can start at the beginning of a shelf and scan the titles. Because librarians have invested a TON of time and effort into classifying the books, you can count on finding many documents about the same topic stored in the same location. There've been efforts to classify the web, but so far nothing really good has popped up. Wikipedia helps in some ways, but it still relies heavily on searches. The contextual navigation from one article to another helps a little, but a lot of the time the articles are linked to one another simply based on the words appearing in the article rather than on whether the articles are strongly related to one another. It does promote serendipitous discovery of information, but it's not so good for finding out a comprehensive list of what's available.
We aren't going away any time soon. Plenty of change a' comin', I reckon, but we're going to be around for a while yet.
[1] Well, technically, I'm a librarian-in-training. Close enough, though.
It's an interesting experiment, but as I said in the comments on his blog post, there's one thing I would change. He says "I'll say I managed to get an illegal copy off a Polish guy I know over the internet." (Emphasis added.)
Why bother lying about it? Just say "I managed to get a copy off a Polish guy I know over the internet." Don't say anything at all about its legal status and allow the listener to assume that it is illegal. There is no need to tell an explicit lie in this case.
And doing so could potentially have adverse consequences. For one thing, the listener may actually have heard about Linux before, and call him on the lie.
For another, it might have personal consequences. It sounds like this a completely informal experiment, and that the guy will be trying it out on his friends. If I was his friend, and I discovered later that he lied to get me to try something, I'd probably be upset with him. Even if the lie meant that I had actually not committed a crime at his suggestion.
Even if telling the lie doesn't have either of those consequences, it explicitly gives them the false idea that Linux cannot legally be copied. So does misdirection of the kind I have proposed, of course; but I tend to think it's easier to correct a mis-impression than an outright lie.
As for the price, I think there probably are people selling Linux for $500 in the States. Certainly there are small vendors selling pre-installed Linux systems for that or more. So that part's not a lie.
The reason that green text on a yellow background is NOT a good idea is that certain types of color blindness (namely protanopia) make it difficult or impossible to distinguish between yellow and green. People with that type of color blindness would perceive it as yellow-on-yellow. Yuck.
Regarding color and accessibility: color can be used to convey meaning; but it's bad practice to rely on color by itself. So, for example, if you have an error message it's fine to give it a light red background, provided that you also provide a non-color-based indicator of what it is (for example, an exclamation point, a stop sign, or similar).
And, for the sake of gratuitous pedantry, the relevant legal requirements regarding web accessibility are not in the ADA (Americans with Disability Act). They're in section 508 of the Rehabilitation Act of 1973, as amended in 1998. The Web Content Accessibility Guidelines 1.0 are also relevant, but do not have the force of law. The Web Content Accessibility Guidelines 2.0 have just gone to last call for comments status after years of work, and will probably be approved by the W3C soon. WCAG 2 is going to be hard to wrap your head around. The standard was designed to be technology independent, and so it's all very abstract.
Slashdot Mirror
I have mixed feelings about Tornquist. He created The Longest Journey which was absolutely amazing, particularly for its story. My favorite game ever. Vivid, detailed characterization, intricate world-building, compelling plot. The tech wasn't impressive (3D figures superimposed on 2D backdrops), but the story was so great that I didn't care.
Then came the sequel, Dreamfall. Oh. My. God. The game was a lot prettier, a good deal more tech glitz. But the UI was atrocious (horrible camera control, unplayable on PC without a USB controller), gratuitous fighting scenes built in (complete with lousy combat controls), and the puzzles (such as they were) didn't make sense. Worse, the plot was incoherent at many crucial points, and the main character (Chloe) completely failed to engage my sympathy or even interest. I got to the end and was sorry she hadn't died permanently somewhere along the way.
Dreamfall had the most severe case of sequel-itis I've ever seen. The original was amazing, astounding, wonderful, and sold a bazillion copies. Then the corporate types took over and threw a ton of cash at the sequel, and it sucked hard. The only comparable thing I can think of? Indiana Jones -- Raiders of the Lost Ark was terrific, and Temple of Doom sucked so hard that nobody ever plays it on TV, not even at 4 AM to fill up time. That's how Dreamfall was.
I have hopes for the third TLJ installment -- after all, The Last Crusade rescued Indiana Jones from one-hit wonder status. It could happen again. But then I think of how the Matrix series went downhill, and fear.
Under blanket licensing, how do I reward artists with good music preferentially to those who suck?
We could always set up the system to reward artists in proportion to 1) the number of times their songs are downloaded, and 2) the average rating of the song.
So:
Step 1: Assess blanket fee (ie tax)
Step 2: Establish a way to track the number of times a song is downloaded and/or played. This could be either a central repository tracking downloads or a distributed system that reports "+1 play Nobody Cares by Levinhurst" to a central server. Obviously lots of potential for abuse if the central server isn't secured properly.
Step 3: Track user ratings for each song. As a simplistic example, scale of 1-5 stars.
Step 4: Work out a formula based on that data, e.g. (base rate * average rating) * total number of plays, calculated at intervals (e.g. monthly).
Step 5: Profit! For the artists, that is.
So if, for example, the base rate is 0.02 (two tenths of a cent) and a single song gets an average rating of 2.3 and plays 201,435 times in April, then sometime in early May the artist gets a payment of (0.02 cents * 2.3 rating) * 201,435 plays = $9,266.01. The minimum rating (e.g. for a song that hasn't been rated by anyone yet) would be 1, to leave the base rate untouched. An artist with a really popular song, highly rated song would get more cash, for example (0.02 cents * 5) * 1,000,437 plays $100,043.70, cha ching! Whereas an artist with a lemon would get fairly little (0.02 cents * 1 rating) * 57 plays = $1.14, doh.
Obviously there are all kinds of details that would have to be worked out. Would the base rate be indexed to inflation? Who would be responsible for allocating the funds? What measures could be taken to prevent crappy artists from using bots to artificially inflate their download/rating? What would an appropriate base rate be? What could we do to account for cost-of-living differences between cheap places and expensive places?
And so on. But don't assume that blanket licensing means everyone gets paid exactly the same. There are lots of ways it could work, and dividing the pot equally amongst all comers is pretty much the dumbest approach.
Professional apps (CAD/simulators/visualizations...) make up the majority of the OpenGL market and they have to be supported for decades (no, military or airlines do not buy a new training system every two years ...)
So breaking compatibility is deal breaker.
Hmm. If OpenGL is being maintained primarily for CAD developers and other people who require stable, long-term APIs, then perhaps the people who maintain the API should explicitly announce that as their goal.
I'm not trying to be snarky. If the needs of game developers are so incompatible with the needs of CAD developers, why not declare OpenGL for CAD and start a new group to focus on the needs of game developers?
There'd probably be a lot of resistance to that, though. Not to mention the many difficulties involved in assembling a group and producing a spec that the developer community (and their employers) can agree on. But it might be worth it.
[1] A fragment is a fancy name for a voxel defined in clip space. After shading and occlusion, the remaining fragments become rasterised as pixels. Thus, the term 'pixel shader' is rather inaccurate.
O_O
I suppose I shouldn't be surprised when people start speaking in alien languages ... I mean, it's Slashdot!
Or perhaps you can enlighten me with a use case for a hobbyist website that requires SSL.
Hi! I have a domain which I acquired solely for the purpose of supplying myself and my family with webmail. I didn't want to have to put up with crappy advertisements, even the politely unintrusive ones in Gmail.
I'd really like to encrypt the log-in sessions for my webmail client. I understand that the email itself is not going to be encrypted once it's sent out. But as it stands, my EMAIL PASSWORD is being transmitted in clear text every time I log in. That sucks. Anyone on any server between me and my web host can get my email password, and they don't even have to go to the trouble of setting up a Man in the Middle attack, because it's completely 100% insecure.
In order to rectify this, I would need to:
All that just to secure my email password? Screw that. I'll keep doing it insecurely and hope to hell nobody hijacks my email to send out spam.
... virtually immediate obsolescence of any printed workFor example, I'm a medievalist. The people I study have all been dead for centuries, and genuinely new data are rare. Once every few years somebody will find a lost manuscript or something, but for the most part we're working with known, thoroughly studied information. Our research doesn't churn; it accretes. I routinely consult articles that are decades old, and in one instance I can think of, I actually cited an article that was over a hundred years old. New research is important too, but it tends to take the form of a new angle on existing data.
Other fields have their own tempos, I'm sure. It's a mistake to assume that all academic fields work alike.
ZDNet is a huge company, and I'm quite sure they've got a staff of full time web developers. It seems rather doubtful that the author of the article had ANY say whatsoever in the site design.
Which is a pity, because I bet he'd do a better job.
I took some measurements. First, some details on my setup. I'm using Firefox (2.0.0.14, because Firebug isn't working smoothly in 3.x yet). I have my browser maximized. I'm using a monitor with a resolution of 1050x1680 (that's a widescreen monitor rotated into portrait orientation so I can read more of a web page at once).
Under those conditions, the web page comes out at 1034 pixels wide by 2985 pixels tall (3,086,490 pixels total). The actual content of the article, which I measured using the Web Developer Toolbar's ruler function, fits into an area of 450 by 860 pixels (387,000). From which we can work out that only 12.5% of the page's area is being used for the actual article. The remaining 87.5% of it is given over to navigation, branding, and advertisements.
Navigation is important, branding is okay, and advertisements are tolerable in small amounts. But in this case, they've totally overwhelmed the actual content of the page, which is stupid. An article of this length should be split into no more than 3 pages, tops. 2 pages would be more reasonable, and it could be done in one page without undue strain, especially since a lot of the article consists of screenshots which don't need to be read closely.
If I read ZDNet on a more regular basis, I'd probably fiddle with their design. A judicious application of "display: none !important" and some other CSS in userContent.css would clean it right up; the at-rule @-moz-document domain(example.com) lets you target a particular site and apply your own CSS rules to their design any time you visit. The !important keyword makes your own rules more important than theirs. I don't do it for many sites, because it takes time and effort. But for the ones I care enough to read often (Salon.com, and, ahem, Slashdot) it's worth it.
In addition to that excellent point, I'm skeptical about the way the table was designed. There's an image of the table here:
http://alistapart.com/d/zebrastripingdoesithelp/data-table.png
The "ordinary" rows have a background color of pure white. The "striped" rows have a background color of #F5F5F5, a very light grey. I'm all in favor of subtle design, but there is such a thing as being too subtle.
Perhaps the stripes did not help noticeably because there was insufficient contrast between the rows?
In the same way, the label "Filesystem" isn't descriptive to a non-techie. She found the search interface, but didn't recognize that label as the one for searching everywhere. How about "Entire Hard Drive" or "All Hard Drives" or, as the article itself suggests, "Whole Computer" instead?
Indexing the partitions, migrating data, all that is potentially useful too, and may well be worth pursuing. But it's still a lot more complex than just picking some friendlier, more descriptive labels and adding a teeny bit more logic to the installer to apply them correctly.
I was very impressed with his work on Pan's Labyrinth, too.
I do have one reservation, though. Del Toro is primarily known as a director of horror films. The vast majority of his work is pretty seriously dark and violent. There are definitely some dark moments and some scary/violent scenes in The Hobbit (such as: the troll attack, riddles in the dark with Gollum, spiders in Mirkwood, and of course the Battle of Five Armies). But there are also a lot of light, delightful scenes (such as: songs in Rivendell, lunch with Beorn, seeing butterflies above Mirkwood, the kindly reception at Lake Town, and so on).
I may be going out on a limb here, but the overall tone of the book slants more towards "delightful" than "scary". Del Toro has amply demonstrated that he can do "scary". But can he do "delightful" just as well? If he can, we're in for a treat. If not, well, who knows what it'll be like? I'll definitely be interested to see what he comes up with; I just hope he does justice to the pleasant stuff as much as the unpleasant stuff.
Just yesterday Marketplace conducted an interview with the Craigslist CEO. You can listen to the interview on their site (it's the "listen to story" link, not the "listen to show" link). It's a pretty interesting interview -- one of the questions was "why aren't you running advertising?" And the answer was "because it's annoying and we think annoying our customers is stupid." When I heard him say that, I sat back and thought "Yeah! Nice to hear a CEO who really gets it."
They're a profitable company; but not madly, hugely, enormously profitable the way they could be. I like that. Capitalism offers powerful incentives to participate in the economy, which is generally good; but it also has a tendency to encourage empire-building and delusions of grandeur. So it's nice to see a company once in a while which isn't hell bent on world domination.
I wish their pages were a bit better designed, though. I appreciate the focus on navigation, but would it kill them to put borders around some of the boxes full of lists, to visually group them? And they could make some improvements for screen reader users, especially adding headings so that blind visitors could have their screen readers jump to the headings in the page rather than having to wade through everything that comes before the part they're interested in.
... beginning college students typically don't know what constitutes "good research". And they tend to be very trusting, not just of Wikipedia, but of anything on the Internet.
A few years ago I had a student turn in a paper arguing that the drinking age should be lowered to 18. One of the claims the paper raised was that drinking ages are lower in many European countries, and that they have healthier drinking cultures. That's probably true, but the source that the student cited to back up the point was totally inadequate. It was a two paragraph account of German drinking habits. The account was based on an interview with an unnamed exchange student. It was written down by an anonymous high school student. And it was put up on the web as a really badly designed web page. Let's see - anonymous author, anonymous interview subject, obviously done as part of a high school assignment, very short, no details, and badly presented. Not exactly the world's most credible source. I made the student go find a more thorough account of European drinking habits written by an identifiable human being and vetted by some kind of editor.
That's a fairly typical example. However, I don't think it's anything worth getting upset about. Students have long been overly credulous. Heck, people in general are overly credulous. It's always been possible to go out, find crappy information, and blindly accept it. Wikipedia (and more broadly the Internet) just make that easier. Yes, there's a lot of GOOD info out there on the web, too, but finding it can be very difficult.
That being the case, I try to integrate assignments about how you do research, and what constitutes a good source, what Internet sources are good for, and when you might want to hit the library and dig a little deeper. It's really a necessity. The students don't know how to do research; therefore, we need to teach them. Many schools are beginning to recognize this -- over the last ten years or so the number of positions at academic libraries for "instructional librarians" has skyrocketed. They visit other teachers' classes and teach lessons on search techniques, evaluation of sources, give tours of the specialized databases the university subscribes to, and so on. Some schools are even beginning to offer complete courses on information literacy. I think we'll probably see a good bit more of this over the next few years.
Andrew Ainsworth's contract does not; he could have requested a share, but he chose not to. His own dumb fault. There's a more informative article over at the Telegraph, which has two important bits of info.
1) There was no formal contract. In the absence of a formal assignment, it is unclear who owned the rights to the design of storm trooper armour. You could argue it either way.
2) Ainsworth is a British citizen who lives in Britain, and always has done. Under UK law, the "design right" in the shape of a physical object (which is NOT the same as copyright) expires after just fifteen years. So, in Britain, he's clear to make and sell just as much of the armour as he likes. It's possible that he may not be able to call it "Storm Trooper" armour; Lucas may have a trademark on that name. But he can definitely call it something more generic, e.g. "space trooper armour" or "space armour" or "sci-fi armour" or something.
Note that this is design right, not copyright. UK law recognizes a difference between the two. Design right applies to the shapes of things.
I'm interested in Mr. Ainsworth's defense. According to the article, he has two claims in his defense. First, "... that the intellectual property rights to the designs have expired." The Copyright, Designs, and Patents Act of 1988 specifies that design right (as distinct from copyright) lasts 15 years from when the design was first recorded or a finished product first made. Since storm troopers appeared in the very first Star Wars in 1977, and had to have been made in 1976 for the filming, I'd say that's a pretty good defense.
The article also says he claims that "if they [the rights] do still exist, as the designer he [Mr. Ainsworth] owns them rather than Mr Lucas." That seems rather more dubious to me. Mr. Ainsworth accepted a payment of 30,000 pounds for his work. Even in the absence of a formal contract, that would strongly suggest that he did it as a commission, in which case ownership of the rights would fall to the commissioner (Mr. Lucas in this case).
It'll be interesting to see how it comes out; and for once we won't have to wait too long, since the case is expected to take ten days.
Nuts. Posted before reading the whole thing - that's got some analysis highlights in it, but mostly it's a list of larger reports you can purchase from Gartner if you want.
The linked blog article is okay if you want a summary, but if you'd prefer, you can check out the complete document. Here's a PDF link to Gartner's full analysis: The State of Open Source 2008.
Interesting. I'm sure the marketers are pleased. The less conscious you are of their message, the less capable you are of resisting it.
Have you considered putting a table right next to the copier?
...
Alternatively, if there's one already there, have you put coasters on it, as a hint?
And if it's got coasters already, have you considered purchasing a cheap mug, drinking coffee out of it just once so it'll have an authentic ring-stain in the bottom, and then setting it on one of those coasters permanently as an added hint?
Failing that, have you taken a bunch of tennis balls, cut them in half, duct taped them to the top of the copier and spray painted them the same beige as the rest so there's no flat place to put drinks?
Further, have you considered sneaking into their cubicles by dead of night and supergluing their cups and mugs to the desk?
If all else fails, have you considered supergluing your coworkers themselves to their desks? I bet their productivity would go up. The smell might get bad after a while, though
Fair enough. ^_^
I've RTF preprint too, and I'm not sure this is a fair criticism. You've misunderstood something - specifically, you write:
Which is incorrect. The researchers chose the brand before the pre-test because they themselves believed it to be associated with creativity. The purpose of the pre-test was to verify that the researchers were correct in believing that the test participants perceive the Apple brand as associated with creativity. They also tested the participants' associations of the IBM logo (which was the other logo they showed) to see whether it too was associated with creativity, and found that it was not.
The end goal of the study was to measure whether or not exposure to a familiar brand would cause people to exhibit characteristics associated with that brand. Having verified that the Apple brand was associated with creativity in the population they were studying, and that the IBM logo was not, they primed their participants by flashing a logo at them for 80 milliseconds as part of a video, then administered a standard psych test measuring creativity (the Unusual Uses test - it was first put together in 1958 and has been thoroughly validated since then). There were 341 participants (190 male, 151 female). Out of those:
They also asked participants after they had completed the test whether they had noticed any images in the video shown beforehand; except for the 80 millisecond flash of logo, it consisted of abstract testing patterns. Not one single participant reported being aware of any recognizable images in the video, so it really was subliminal.
Their analysis of the results showed statistically significant differences between those shown IBM and those shown the Apple logo. Specifically, the people who got Apple produced a greater number of responses on the unusual uses test, which is a quantitative difference that nicely demonstrates their desire to be creative if not the actual creativity. The researchers also had judges independently rate the responses on a qualitative scale. The judges didn't know which logo the participants they were evaluation had seen. The researchers then did statistical analysis on the qualitative judgments, which showed that the judges consistently rated the responses of the Apple participants as more creative than those of the IBM participants.
Then they did it again, two more times in fact, with different brands and traits.
They had lots of checks and counter-checks built into it, and they used some fairly sophisticated statistical analysis on the results, including some analysis of variance (ANOVA) checks. As is common with psychological studies, the test participants were all undergraduates taking an intro to psych class. So the population was not particularly diverse in terms of age. But on the whole, it looks to me like a pretty well designed study.
Yes, that's true. Human judgment about what constitutes high quality information helps a lot.
... that may well be very interesting and useful to local historians and genealogists from Chillicothe and surrounding towns. But it's not really going to have a large enough market for a commercial publisher to be interested. So instead you print up, say, 30 copies with a vanity press, and distribute them to local libraries and churches.
Regarding self-published material - yes, it's rare in libraries. I would add though that the fact that something was self-published doesn't necessarily mean that it's low quality. It could just mean that it's not commercially viable. Example: genealogy. If I write a genealogical history of families sharing the last name "Potter" who settled in Chillicothe, Virginia, during the last 150 years
Hi, I'm a librarian.[1] I appreciate your response, and I'm glad you find us useful. I'd just like to elaborate on one of your points. You wrote "I like going to the library just to browse and to see what I can find." (Emphasis added). This is one point where physical libraries still have a distinct advantage over the Internet.
People who are trying to retrieve information have three basic types of queries.
The Internet is pretty good for known-item searches. Especially if the item has indexable text in it. Other types of information are harder. Quick! Using Google Images, find me a picture of a sheep facing left at sunset.
The Internet is less good at delivering focused results for a known class search. It can retrieve relevant documents, but there's a good chance that it will also retrieve lots of unrelated or only tangentially related things. Which means you have to spend ages sorting through a giant list of search results to find what you really want. Specialized databases tend to produce much more focused results, of course, but most of those aren't freely available.
And lastly, the Internet is lousy for browsing. Browsing is about finding out what's available within a very broad class of stuff. Search engines can tell you that documents share keywords; they can't tell you for certain that the documents are actually about similar things. And within the search results, they're organized according to (roughly) how popular they are, as measured by how many sites link to them. They're not organized based on their similarities to or differences from one another. Compare to a library, where you can start at the beginning of a shelf and scan the titles. Because librarians have invested a TON of time and effort into classifying the books, you can count on finding many documents about the same topic stored in the same location. There've been efforts to classify the web, but so far nothing really good has popped up. Wikipedia helps in some ways, but it still relies heavily on searches. The contextual navigation from one article to another helps a little, but a lot of the time the articles are linked to one another simply based on the words appearing in the article rather than on whether the articles are strongly related to one another. It does promote serendipitous discovery of information, but it's not so good for finding out a comprehensive list of what's available.
We aren't going away any time soon. Plenty of change a' comin', I reckon, but we're going to be around for a while yet.
[1] Well, technically, I'm a librarian-in-training. Close enough, though.
It's an interesting experiment, but as I said in the comments on his blog post, there's one thing I would change. He says "I'll say I managed to get an illegal copy off a Polish guy I know over the internet." (Emphasis added.)
Why bother lying about it? Just say "I managed to get a copy off a Polish guy I know over the internet." Don't say anything at all about its legal status and allow the listener to assume that it is illegal. There is no need to tell an explicit lie in this case.
And doing so could potentially have adverse consequences. For one thing, the listener may actually have heard about Linux before, and call him on the lie.
For another, it might have personal consequences. It sounds like this a completely informal experiment, and that the guy will be trying it out on his friends. If I was his friend, and I discovered later that he lied to get me to try something, I'd probably be upset with him. Even if the lie meant that I had actually not committed a crime at his suggestion.
Even if telling the lie doesn't have either of those consequences, it explicitly gives them the false idea that Linux cannot legally be copied. So does misdirection of the kind I have proposed, of course; but I tend to think it's easier to correct a mis-impression than an outright lie.
As for the price, I think there probably are people selling Linux for $500 in the States. Certainly there are small vendors selling pre-installed Linux systems for that or more. So that part's not a lie.
The reason that green text on a yellow background is NOT a good idea is that certain types of color blindness (namely protanopia) make it difficult or impossible to distinguish between yellow and green. People with that type of color blindness would perceive it as yellow-on-yellow. Yuck.
As for Nielson's ugly site, that's deliberate.
Regarding color and accessibility: color can be used to convey meaning; but it's bad practice to rely on color by itself. So, for example, if you have an error message it's fine to give it a light red background, provided that you also provide a non-color-based indicator of what it is (for example, an exclamation point, a stop sign, or similar).
And, for the sake of gratuitous pedantry, the relevant legal requirements regarding web accessibility are not in the ADA (Americans with Disability Act). They're in section 508 of the Rehabilitation Act of 1973, as amended in 1998. The Web Content Accessibility Guidelines 1.0 are also relevant, but do not have the force of law. The Web Content Accessibility Guidelines 2.0 have just gone to last call for comments status after years of work, and will probably be approved by the W3C soon. WCAG 2 is going to be hard to wrap your head around. The standard was designed to be technology independent, and so it's all very abstract.