I haven't read the article yet, because I'm going to try a test. Does the article say anything about false alarms? Because that's probably the most important thing we need to know about this scheme.
Will it go off when one of those unmarked white trucks that's used for discreet transport of nuclear waste goes by? How about when the big research hospital gets a shipment of isotopes for cancer treatment? How about shipments of nuclear weapons by the military?
It's quite possible that such a system might reveal quite a bit of classified information and have all sorts of unintended consequences. A few years back there was a flap because the U. S. had sent a ship into a Japanese port which the Japanese suspected contained nuclear weapons. As always, the U. S. refused to confirm or deny the fact. But if cell phones were organized into radiation detection networks, then not only would it unmask military secrets, but it might trigger mass panic in the interval between the nuclear material's being detected and the military reluctantly confirming that that's what it was.
OK, let's see if the article talks about false alarms at all:
Nope, just as I thought. It is devoted entirely to how sensitive the system is and how it can detect weak radioactive sources. Not a single sentence about how common innocent weak radioactive sources are, or what the distribution of weak radioactive sources is like when there are no terrorists around.
Do we see a pattern here? Not just with this administration, but in general.
Some authority engages in controversial, borderline activity that might be illegal. It transpires that the activities were recorded (taped, logged, written in memos). Investigator tells entity to save those records. The mills of justice grind slowly. It then transpires that the records have been shredded, deleted, bulk-erased, recycled, whatever.
Authority's spokeperson smirks*. Everybody knows darn well that the destruction was deliberate, but everybody knows darn well that there's absolutely no way to prove it.
Nobody even needs to tell subordinates what to do in any detail. In many cases, all that's needed is to do nothing. It takes exceptional action to stop the janitor from emptying the wastebasket, stop the operator from reusing the tapes, whatever.
I don't think there's a thing to do about this sort of stuff. But I just hope that once, just once, one of the bastards gets taped in the act of ordering the destruction of those tapes, and--
I'm just stunned at someone coming up with a totally new way to do something simple (hold liquid) in a simple way (in a container of the right shape) based on a familiar principle (surface tension).
In a sense, the idea of using surface tension to hold fluids is not new--think of a sponge or a towel--but getting cup-like and pipe-like functionality is.
I've no doubt that if humans had evolved in zero gravity this would have been discovered back around the same time as clay pots and chipped flint arrowheads, but as it is they didn't.
It's nice to know there are still inventions to be invented that don't rely on a billion microchips and a million lines of code.
Two bitterly competitive rivals both seek a means of becoming invisible.
One of them believes (incorrectly) that if he can find a perfectly black substance (now available!) and coat himself with it, he will become invisible. The other thinks that it should not be too hard to make his body perfectly transparent.
In the story, both methods succeed, but both have a flaw. The black-coated brother still casts a shadow, and when he is around, you can't see him but you nevertheless feel a mysterious "sudden cold chill, reminding me of deep mines and gloomy crypts. The transparent brother evokes rainbow-colored flashes when the light hits him at the right angle.
Read it--the full text, online, is linked above--it's a stitch...
Scroll down in the comments, to where someone named "Bran" (Peter Brando, according to the link" says "I work with MS Professional Support" and comments, apparently with a straight face:
"10 years is definitely a long time to have a case open."
...and will this software be installed on their computers? To monitor their heart rate, body temperature, movement, facial expression and blood pressure?
After all, "stress and frustration" have more serious consequences if they lead to bad decision-making.
How about generals? How about the Commander-in-Chief? Isn't their "productivity, physical wellbeing and competence" important?
Why do I somehow think that it is not going to be installed on any management machines... and that the stated rationales are pretexts?
99% accuracy rate is actually pretty bad in the real world. In a typical document, you might expect 12-15 words per line - so you have one error every 7 lines or so.
Right, and my brief experience trying out ViaVoice convinced me that even that observation underestimates the seriousness of the problem.
1) You don't necessarily notice the errors as you make them.
2) Correcting errors takes a surprisingly large amount of attention and labor, as well as being a distraction from the real task.
3) Correcting errors on the keyboard feels like task-switching and feels distracting and laborious. Correcting errors by issuing verbal commands feels more natural and less laborious... but not infrequently gets you into the situation where the speech recognizer misunderstands your commands, which rapidly snowballs into a real mess.
It all reminds me of a situation some twenty years ago. In the company I worked for, typically, non-native-English-speaking scientists would typed up their papers and hand them to an "editorial services" department, which began by rekeying them into a word processor. I was present when some sales representatives were making a pitch for OCR equipment to the department head. She listened politely until they said that the equipment was "99.5% accurate." She stopped them right there.
"If you can't guarantee 100% accuracy," she said, "I'm not interested. Almost all the labor in rekeying is incurred, not in the actual keying, but in the subsequent proofreading. A document that's almost error-free takes just about as long to proofread as one that has a normal number of errors. The only thing that would give us a significant labor savings would be a system so accurate that we could eliminate the proofreading step."
I'm convinced that although the durability of product features is almost entirely governed by the time horizons of employees within companies.
The only way you get consistent backward compatibility and a consistent style is when the product is being developed and managed by a consistent set of people.
So much of the important stuff is in peoples' minds and hearts.
You can embed the important stuff on paper, of course, with standards and style guides. But people only follow them... to the spirit, not to the letter... if the people working on them have bought in and care about them. And, of course, if managers who decide what standards and style guides to use keep them in place.
This is simply taking advantage of mom 'n pop consumers who are just out to buy a nice birthday gift or something like that and don't read consumer electronics news sites.
There's probably nothing in particular that can be done to stop it. It's simply the strong taking from the weak, where in this case the weak are the uninformed. The current moral climate in the United States seems to accept that it is perfectly OK for the strong to take from the weak as long as there's no law against it, and as long as it only involves money. But it leaves a bad taste in my mouth nonetheless.
I wonder how many of the Best Buys of the world will be warning customers that the price drop is a firesale of a product that many think will be orphaned, and how many will be stacking 'em up by the checkout isles and selling them as hard as they can?
...I'm afraid I don't recall the brands, but several makers of video terminals used layouts that inserted an extra key in the bottom row, thus placing the CTRL key one key-width farther left than usual. Of course that required relearning--whenever I used one of those keyboard, for the first half-hour or so I'd keep hitting the extra key when I meant to hit CTRL, but that wasn't the problem.
The problem was that every CTRL combination required you to stretch your pinky that much further from the rest of your fingers than usual.
And one of them was at a company that used emacs as their standard text editor.
That was the only time in my life that using a computer made my hands, or rather my left hand, hurt so badly that I was on the verge of seeing a doctor. I trained myself to type all CTRL combinations using two hands, and the problem gradually subsided.
...or, in my case, the VIC-20, virtually identical.
The nonstandard layout criticism shouldn't apply, because in those days there were no standards for video terminal keyboards or computer keyboards.
The knock on keyboard height is legitimate but overstated. It was about the same height as other video terminal keyboards in its day. The Europeans instituted ergonomic regulations that resulted in very slim, low-height keyboards we're familiar with, but they didn't really start to take hold until, say, 1980 or so.
At the time, I regularly used numerous computer terminals at work, including most major brands: DEC's VT100, the $5000 built-like-a-tank-no-corners-cut HP2648A, the LSI ADM3A, and many others; and I spent a lot of keying time on the Apple ][.
The HP2648A happened to be the one that made my hands hurt, because (for some reason) the keys featured a combination of a fairly large travel distance and a fairly stiff spring.
The VIC-20 and Commodore 64 had a very nice feel to them and were very easy on the hands, apart from keyboard height.
Keyboard height in itself was not a problem if you held you hands correctly (which I did) or used a wrist rest.
Re:They didn't make it for you
on
Negroponte vs Intel
·
· Score: 2, Insightful
Regrettably, I agree with the original poster.
It may not matter. The Sugar UI is an adequate application (sorry, "activity") launcher and the applications themselves are individually good.
There were many valid reasons for building a completely new UI from the ground up, and many of the unfamiliar and "different" ways of doing things are the result of an attempt to meet different needs, and originality.
Still, the implementation is currently a mess and simply does not achieve the visions articulated in the style guide.
For example, "recoverability" is explicitly enunciated as a key design principle, yet there is no "undo" function in most contexts, as far as I could tell, anyway. (For example, if you delete a Journal entry, it's gone).
The idea of the Journal and dispensing with hierarchies sounds good in principle, but in practice I certainly have trouble finding stuff. It appears to be necessary to type in tags and descriptions if you want to be able to find them with the Search feature.
The OLPC documentation tapdances around this by saying that "fortunately" describing things is a natural activity for children. Fine, but does this mean it is natural for a seven-year-old to type in descriptions into Journal fields?
One of the highly touted features of the XO is its ability to display the source code for all of its own programs ("activities.") It doesn't work. The button (actually a key combination, CTRL-U or FN-Space (FN-Gear-key)) does nothing.
Faulty DRM and "software activation" schemes are logic bombs, too.
There is of course a a very important difference, in that they are not intended to do anything but enforce the bombers' legal rights. Or, at any rate, what the bombers credibly believe to be their legal rights.
But when a malfunctioning Microsoft server trips the "kill" switch on legitimate copies of Vista, I think it's fair to call that a logic bomb of sorts.
No, I don't think Bill Gates should do 2.5 years of jail time, but it is disappointing that Microsoft was not held accountable for this beyond a few weeks' of mildly embarrassing publicity.
The story I'm waiting for is a story about a media company spokesperson saying:
"We've determined that most viewers don't care about the difference between either HD format and and standard DVD, so we have decided to drop both HD-DVD and Blu-Ray, and from now on will issue all of our releases in DVD only. We save money by not having to support the new formats, and, accordingly, we are lowering the prices of all our DVDs by 15%."
"The use of a signal to mimic a missile attack has already been tested in the air, said Tim Wagner, an American Airlines spokesman." Yeah, right. So they're not going to test it with real missile, which doesn't give a lot of confidence that it will actually work.
Sounds like that "successful" antimissile test they did a year or so ago, where the missile was conveniently equipped with a GPS unit that continuously radioed its position to the antimissile system.
On the other hand, are they going to use signals to "mimic" things that are not missile attacks... like near-miss encounters with other passenger jets, for example?
"Burt Keirstead, director of BAE's commercial airline protection program, said BAE's contract requires it to prove that Jeteye will operate without failure for 3,000 hours of flight, and sets a goal of 4,500 hours."
What constitutes a failure? If it shoots at a Medivac helicopter and brings it down, did it succeed or fail?
And why should there be such a thing as a "digital issue?"
I don't remember any "modulation issues" when FM radio was introduced. You just bought the damn radio and it worked, except the sound was better than AM.
I don't remember any "magnetic issues" when cassettes were introduced. You just bought the damn cassette player and it worked, except the sound wasn't quite as good as LPs... but the cassettes were compact and there weren't any ticks, pops, or scratches.
And for that matter I don't remember any "digital issues" when CDs were introduced. You just bought the damn CD player and it worked, except that the sound was better then on cassettes. (And for 98% of all ears on 98% of all recordings in 98% of all real-world consumer situations, it was much better than LPs, too).
If the customer is using HDMI and having "issues," then they're HDMI issues.
So, for the last five years we've been watching Microsoft drop most of the important stuff from Vista, on the premise that by doing so they'd get a good, clean release out of the door.
The trade press looks at... what was that called? It was not a beta, I believe it was called a "release candidate..." and everyone says, "Gee, there's really some serious suckage here." Strike one.
But the Microsoft advocates say, "Whoa, it's not the release, it's just a release candidate, it's not fair to judge it, they'll get that all cleaned up for release.
So, it's released, and the trade press looks at it, and everyone says, "Gee, there's really some serious suckage here." ("Greatest tech disappointment of 2007" is how PC World puts it). No excited early adopters running into work and saying "Oh, boy! I just bought a new laptop. You've gotta see Vista. Aero is so cool. And it's so fast! I just love it." Strike 2.
But the Microsoft advocates say, "Whoa, everyone knows there are teething pains with the first release. It doesn't mean a thing. Sophisticated buyers and corporations know that it's SP1 that matters."
SP1 was Microsoft's second chance to make a good first impression.
So, is there anyone out there saying "It doesn't matter, it's not a big deal, SP2 is what really counts?
Another chess move in a long, complicated game that is either being played by such idiots that it will last a long time, because none of them knows how to play... or is being played by such subtle, clever, brilliant chessplayers that it will last a long time, because none will be able to get a commanding advantage over the others.
Either way, I can't predict the winner, so why should I care?
I agree with all of your judgments, and am relieved to see I'm not alone.
The UI is not awful, and is good enough, and it was probably correct to think it through from scratch instead of trying to riff on the Alto/Star/Lisa/Mac/Windows. But it still tastes to me like other not-so-good UIs, in which the designer and people that can be coaxed into the same mindspace can be convinced that it's better than it is.
I read the human interface guidelines and I'm not convinced. I've often talked to people who have believed their UI was easy to use because "you always do thus-and-such to achieve this-and-that, and the frammises are always on the left edge, and you ferthboinder toward the top to glorp persistent quibinicks..."
One of the things that was fascinating about the Mac in 1984, which I approached with virtually no previous experience, was that you could intuit it and use it without ever formulating or deducing the consistent left-brained rules by which it operated. For about three days I used it effectively without understanding it at all. I wanted to achieve something, I took a wild guess as to what might work, and it usually did.
I don't feel that way about Sugar, although maybe my brain has just ossified.
If the Journal functioned the way it's supposed to, I don't understand why it, rather than the "home view," isn't the center of the user experience, and the thing you boot into. Seems to me that you'd more often be returning to an old activity than starting a new one.
I "get" the idea of a linear, chronological arrangement of activities rather than a hierarchical tree of documents, but I don't understand how you navigate that arrangement unless you are punctilious about giving each saved activity a good name, and clever at naming them in such a way that you can search for them by typing search strings (which I think only search the name of the journal entry, not the content of the saved activity).
I'm glad to hear this. I don't want to press you on this, but just how literally true is it that you didn't give them any instruction? Are you sure you didn't give them a hint or a tip from time to time... or go online into the OLPC site to their user's guide in order to figure out the Journal, Neighborhood, Home keys did and stuff like that?
I interpret what you said to mean that they were able to guess or discover the meaning of the icons for the browser, word processor and start using them without help.
I also interpret this to mean that they have not mastered the Journal, which is one of the UI elements I find troublesome. I don't object in principle to organizing things by activity and time and linearly, rather than in the now-traditional filesystem hierarchy, but apparently in order to get back to the stuff you've worked on you need to a) remember to give it a good name and b) type a good search string in the Journal.
People can't figure your OS out because the menu commands keep moving into new menus with each release and the toolbar icons are too small to represent anything? Add context-sensitive help.
The context-sensitive help requires you to work through a clumsy twenty-step process to achieve something? Add "wizards" which force you through the twenty-step process, one slow, painful step at a time. (Converting Xerox Alto's "modeless" paradigm into a good old IBM 704 paradigm...)
People can't figure out how to pop up the "wizards?" Add wizards that pop up by themselves (Clippy).
People get frustrated? Add an automatic frustration detector.
How the heck can you satirize Microsoft when they do such a good job of satirizing themselves?
Slashdot Mirror
I haven't read the article yet, because I'm going to try a test. Does the article say anything about false alarms? Because that's probably the most important thing we need to know about this scheme.
Will it go off when one of those unmarked white trucks that's used for discreet transport of nuclear waste goes by? How about when the big research hospital gets a shipment of isotopes for cancer treatment? How about shipments of nuclear weapons by the military?
It's quite possible that such a system might reveal quite a bit of classified information and have all sorts of unintended consequences. A few years back there was a flap because the U. S. had sent a ship into a Japanese port which the Japanese suspected contained nuclear weapons. As always, the U. S. refused to confirm or deny the fact. But if cell phones were organized into radiation detection networks, then not only would it unmask military secrets, but it might trigger mass panic in the interval between the nuclear material's being detected and the military reluctantly confirming that that's what it was.
OK, let's see if the article talks about false alarms at all:
Nope, just as I thought. It is devoted entirely to how sensitive the system is and how it can detect weak radioactive sources. Not a single sentence about how common innocent weak radioactive sources are, or what the distribution of weak radioactive sources is like when there are no terrorists around.
...that you have to add a tag to the page to say that it is coded to the standard?
If the standard says that a page will render properly without that tag, then IE8 is not standards-compliant. It's as simple as that.
And a W3C truth squad should hold them to it, and complain whenever and whereaver Microsoft claims that IE8 complies with the standard.
Won't someone at least hint at who those officials were, so that I can start making my ideological prejudgments on the credibility of the allegations?
Do we see a pattern here? Not just with this administration, but in general.
Some authority engages in controversial, borderline activity that might be illegal. It transpires that the activities were recorded (taped, logged, written in memos). Investigator tells entity to save those records. The mills of justice grind slowly. It then transpires that the records have been shredded, deleted, bulk-erased, recycled, whatever.
Authority's spokeperson smirks*. Everybody knows darn well that the destruction was deliberate, but everybody knows darn well that there's absolutely no way to prove it.
Nobody even needs to tell subordinates what to do in any detail. In many cases, all that's needed is to do nothing. It takes exceptional action to stop the janitor from emptying the wastebasket, stop the operator from reusing the tapes, whatever.
In the Boston area there is a controversial school, the Judge Rotenberg Center, which uses electric shocks to train kids with behavioral problems. Recently, a kid at the center who had not done anything disruptive was subjected to a long series of shocks, on the basis of telephoned instructions from a "prank" caller. The shock treatment was taped. State investigator ordered the center to preserve the tapes. Surprise, surprise: they were destroyed. Because, in the opinion of the head of the Institute, the investigation "seemed to be finished."
I don't think there's a thing to do about this sort of stuff. But I just hope that once, just once, one of the bastards gets taped in the act of ordering the destruction of those tapes, and--
--destroys that tape too?
Oh well, never mind.
*OK, I'm just imagining that smirk.
...at least I think it's original.
I'm just stunned at someone coming up with a totally new way to do something simple (hold liquid) in a simple way (in a container of the right shape) based on a familiar principle (surface tension).
In a sense, the idea of using surface tension to hold fluids is not new--think of a sponge or a towel--but getting cup-like and pipe-like functionality is.
I've no doubt that if humans had evolved in zero gravity this would have been discovered back around the same time as clay pots and chipped flint arrowheads, but as it is they didn't.
It's nice to know there are still inventions to be invented that don't rely on a billion microchips and a million lines of code.
The science-fiction story that's really appropriate here is Jack London's story, The Shadow and the Flash. The "science" is cockamamie, but it's amusing anyway.
Two bitterly competitive rivals both seek a means of becoming invisible.
One of them believes (incorrectly) that if he can find a perfectly black substance (now available!) and coat himself with it, he will become invisible. The other thinks that it should not be too hard to make his body perfectly transparent.
In the story, both methods succeed, but both have a flaw. The black-coated brother still casts a shadow, and when he is around, you can't see him but you nevertheless feel a mysterious "sudden cold chill, reminding me of deep mines and gloomy crypts. The transparent brother evokes rainbow-colored flashes when the light hits him at the right angle.
Read it--the full text, online, is linked above--it's a stitch...
Scroll down in the comments, to where someone named "Bran" (Peter Brando, according to the link" says "I work with MS Professional Support" and comments, apparently with a straight face:
"10 years is definitely a long time to have a case open."
...and will this software be installed on their computers? To monitor their heart rate, body temperature, movement, facial expression and blood pressure?
After all, "stress and frustration" have more serious consequences if they lead to bad decision-making.
How about generals? How about the Commander-in-Chief? Isn't their "productivity, physical wellbeing and competence" important?
Why do I somehow think that it is not going to be installed on any management machines... and that the stated rationales are pretexts?
99% accuracy rate is actually pretty bad in the real world. In a typical document, you might expect 12-15 words per line - so you have one error every 7 lines or so.
Right, and my brief experience trying out ViaVoice convinced me that even that observation underestimates the seriousness of the problem.
1) You don't necessarily notice the errors as you make them.
2) Correcting errors takes a surprisingly large amount of attention and labor, as well as being a distraction from the real task.
3) Correcting errors on the keyboard feels like task-switching and feels distracting and laborious. Correcting errors by issuing verbal commands feels more natural and less laborious... but not infrequently gets you into the situation where the speech recognizer misunderstands your commands, which rapidly snowballs into a real mess.
It all reminds me of a situation some twenty years ago. In the company I worked for, typically, non-native-English-speaking scientists would typed up their papers and hand them to an "editorial services" department, which began by rekeying them into a word processor. I was present when some sales representatives were making a pitch for OCR equipment to the department head. She listened politely until they said that the equipment was "99.5% accurate." She stopped them right there.
"If you can't guarantee 100% accuracy," she said, "I'm not interested. Almost all the labor in rekeying is incurred, not in the actual keying, but in the subsequent proofreading. A document that's almost error-free takes just about as long to proofread as one that has a normal number of errors. The only thing that would give us a significant labor savings would be a system so accurate that we could eliminate the proofreading step."
I'm convinced that although the durability of product features is almost entirely governed by the time horizons of employees within companies.
The only way you get consistent backward compatibility and a consistent style is when the product is being developed and managed by a consistent set of people.
So much of the important stuff is in peoples' minds and hearts.
You can embed the important stuff on paper, of course, with standards and style guides. But people only follow them... to the spirit, not to the letter... if the people working on them have bought in and care about them. And, of course, if managers who decide what standards and style guides to use keep them in place.
I really hate moves like this.
This is simply taking advantage of mom 'n pop consumers who are just out to buy a nice birthday gift or something like that and don't read consumer electronics news sites.
There's probably nothing in particular that can be done to stop it. It's simply the strong taking from the weak, where in this case the weak are the uninformed. The current moral climate in the United States seems to accept that it is perfectly OK for the strong to take from the weak as long as there's no law against it, and as long as it only involves money. But it leaves a bad taste in my mouth nonetheless.
I wonder how many of the Best Buys of the world will be warning customers that the price drop is a firesale of a product that many think will be orphaned, and how many will be stacking 'em up by the checkout isles and selling them as hard as they can?
The "view source" key combination doesn't seem to work, at least not on the G1G1 machines, or at least I haven't had any luck with it so far.
It displays the HTML source of a web page, but doesn't appear to do anything at all anywhere else. I'm referring to fn + space (the "gear" symbol).
Too bad, as it was one of the most intriguing features of the XO. I'm not sure quite what happened.
I don't really think Microsoft bribed anyone at OLPC to sabotage it...
...I'm afraid I don't recall the brands, but several makers of video terminals used layouts that inserted an extra key in the bottom row, thus placing the CTRL key one key-width farther left than usual. Of course that required relearning--whenever I used one of those keyboard, for the first half-hour or so I'd keep hitting the extra key when I meant to hit CTRL, but that wasn't the problem.
The problem was that every CTRL combination required you to stretch your pinky that much further from the rest of your fingers than usual.
And one of them was at a company that used emacs as their standard text editor.
That was the only time in my life that using a computer made my hands, or rather my left hand, hurt so badly that I was on the verge of seeing a doctor. I trained myself to type all CTRL combinations using two hands, and the problem gradually subsided.
...or, in my case, the VIC-20, virtually identical.
The nonstandard layout criticism shouldn't apply, because in those days there were no standards for video terminal keyboards or computer keyboards.
The knock on keyboard height is legitimate but overstated. It was about the same height as other video terminal keyboards in its day. The Europeans instituted ergonomic regulations that resulted in very slim, low-height keyboards we're familiar with, but they didn't really start to take hold until, say, 1980 or so.
At the time, I regularly used numerous computer terminals at work, including most major brands: DEC's VT100, the $5000 built-like-a-tank-no-corners-cut HP2648A, the LSI ADM3A, and many others; and I spent a lot of keying time on the Apple ][.
The HP2648A happened to be the one that made my hands hurt, because (for some reason) the keys featured a combination of a fairly large travel distance and a fairly stiff spring.
The VIC-20 and Commodore 64 had a very nice feel to them and were very easy on the hands, apart from keyboard height.
Keyboard height in itself was not a problem if you held you hands correctly (which I did) or used a wrist rest.
Regrettably, I agree with the original poster.
It may not matter. The Sugar UI is an adequate application (sorry, "activity") launcher and the applications themselves are individually good.
There were many valid reasons for building a completely new UI from the ground up, and many of the unfamiliar and "different" ways of doing things are the result of an attempt to meet different needs, and originality.
Still, the implementation is currently a mess and simply does not achieve the visions articulated in the style guide.
For example, "recoverability" is explicitly enunciated as a key design principle, yet there is no "undo" function in most contexts, as far as I could tell, anyway. (For example, if you delete a Journal entry, it's gone).
The idea of the Journal and dispensing with hierarchies sounds good in principle, but in practice I certainly have trouble finding stuff. It appears to be necessary to type in tags and descriptions if you want to be able to find them with the Search feature.
The OLPC documentation tapdances around this by saying that "fortunately" describing things is a natural activity for children. Fine, but does this mean it is natural for a seven-year-old to type in descriptions into Journal fields?
One of the highly touted features of the XO is its ability to display the source code for all of its own programs ("activities.") It doesn't work. The button (actually a key combination, CTRL-U or FN-Space (FN-Gear-key)) does nothing.
Faulty DRM and "software activation" schemes are logic bombs, too.
There is of course a a very important difference, in that they are not intended to do anything but enforce the bombers' legal rights. Or, at any rate, what the bombers credibly believe to be their legal rights.
But when a malfunctioning Microsoft server trips the "kill" switch on legitimate copies of Vista, I think it's fair to call that a logic bomb of sorts.
No, I don't think Bill Gates should do 2.5 years of jail time, but it is disappointing that Microsoft was not held accountable for this beyond a few weeks' of mildly embarrassing publicity.
The story I'm waiting for is a story about a media company spokesperson saying:
"We've determined that most viewers don't care about the difference between either HD format and and standard DVD, so we have decided to drop both HD-DVD and Blu-Ray, and from now on will issue all of our releases in DVD only. We save money by not having to support the new formats, and, accordingly, we are lowering the prices of all our DVDs by 15%."
Now, that would be news.
"The use of a signal to mimic a missile attack has already been tested in the air, said Tim Wagner, an American Airlines spokesman." Yeah, right. So they're not going to test it with real missile, which doesn't give a lot of confidence that it will actually work.
Sounds like that "successful" antimissile test they did a year or so ago, where the missile was conveniently equipped with a GPS unit that continuously radioed its position to the antimissile system.
On the other hand, are they going to use signals to "mimic" things that are not missile attacks... like near-miss encounters with other passenger jets, for example?
"Burt Keirstead, director of BAE's commercial airline protection program, said BAE's contract requires it to prove that Jeteye will operate without failure for 3,000 hours of flight, and sets a goal of 4,500 hours."
What constitutes a failure? If it shoots at a Medivac helicopter and brings it down, did it succeed or fail?
And why should there be such a thing as a "digital issue?"
I don't remember any "modulation issues" when FM radio was introduced. You just bought the damn radio and it worked, except the sound was better than AM.
I don't remember any "magnetic issues" when cassettes were introduced. You just bought the damn cassette player and it worked, except the sound wasn't quite as good as LPs... but the cassettes were compact and there weren't any ticks, pops, or scratches.
And for that matter I don't remember any "digital issues" when CDs were introduced. You just bought the damn CD player and it worked, except that the sound was better then on cassettes. (And for 98% of all ears on 98% of all recordings in 98% of all real-world consumer situations, it was much better than LPs, too).
If the customer is using HDMI and having "issues," then they're HDMI issues.
So, for the last five years we've been watching Microsoft drop most of the important stuff from Vista, on the premise that by doing so they'd get a good, clean release out of the door.
The trade press looks at... what was that called? It was not a beta, I believe it was called a "release candidate..." and everyone says, "Gee, there's really some serious suckage here." Strike one.
But the Microsoft advocates say, "Whoa, it's not the release, it's just a release candidate, it's not fair to judge it, they'll get that all cleaned up for release.
So, it's released, and the trade press looks at it, and everyone says, "Gee, there's really some serious suckage here." ("Greatest tech disappointment of 2007" is how PC World puts it). No excited early adopters running into work and saying "Oh, boy! I just bought a new laptop. You've gotta see Vista. Aero is so cool. And it's so fast! I just love it." Strike 2.
But the Microsoft advocates say, "Whoa, everyone knows there are teething pains with the first release. It doesn't mean a thing. Sophisticated buyers and corporations know that it's SP1 that matters."
SP1 was Microsoft's second chance to make a good first impression.
So, is there anyone out there saying "It doesn't matter, it's not a big deal, SP2 is what really counts?
Another chess move in a long, complicated game that is either being played by such idiots that it will last a long time, because none of them knows how to play... or is being played by such subtle, clever, brilliant chessplayers that it will last a long time, because none will be able to get a commanding advantage over the others.
Either way, I can't predict the winner, so why should I care?
I agree with all of your judgments, and am relieved to see I'm not alone.
The UI is not awful, and is good enough, and it was probably correct to think it through from scratch instead of trying to riff on the Alto/Star/Lisa/Mac/Windows. But it still tastes to me like other not-so-good UIs, in which the designer and people that can be coaxed into the same mindspace can be convinced that it's better than it is.
I read the human interface guidelines and I'm not convinced. I've often talked to people who have believed their UI was easy to use because "you always do thus-and-such to achieve this-and-that, and the frammises are always on the left edge, and you ferthboinder toward the top to glorp persistent quibinicks..."
One of the things that was fascinating about the Mac in 1984, which I approached with virtually no previous experience, was that you could intuit it and use it without ever formulating or deducing the consistent left-brained rules by which it operated. For about three days I used it effectively without understanding it at all. I wanted to achieve something, I took a wild guess as to what might work, and it usually did.
I don't feel that way about Sugar, although maybe my brain has just ossified.
If the Journal functioned the way it's supposed to, I don't understand why it, rather than the "home view," isn't the center of the user experience, and the thing you boot into. Seems to me that you'd more often be returning to an old activity than starting a new one.
I "get" the idea of a linear, chronological arrangement of activities rather than a hierarchical tree of documents, but I don't understand how you navigate that arrangement unless you are punctilious about giving each saved activity a good name, and clever at naming them in such a way that you can search for them by typing search strings (which I think only search the name of the journal entry, not the content of the saved activity).
I'm glad to hear this. I don't want to press you on this, but just how literally true is it that you didn't give them any instruction? Are you sure you didn't give them a hint or a tip from time to time... or go online into the OLPC site to their user's guide in order to figure out the Journal, Neighborhood, Home keys did and stuff like that?
I interpret what you said to mean that they were able to guess or discover the meaning of the icons for the browser, word processor and start using them without help.
I also interpret this to mean that they have not mastered the Journal, which is one of the UI elements I find troublesome. I don't object in principle to organizing things by activity and time and linearly, rather than in the now-traditional filesystem hierarchy, but apparently in order to get back to the stuff you've worked on you need to a) remember to give it a good name and b) type a good search string in the Journal.
She: I'm perfect! ...I think I saw this in Voo Doo, MIT's humor magazine, in the 1960s, but it's probably much older than that.
He: I'm practice!
Don't fix anything, just apply bandaids.
People can't figure your OS out because the menu commands keep moving into new menus with each release and the toolbar icons are too small to represent anything? Add context-sensitive help.
The context-sensitive help requires you to work through a clumsy twenty-step process to achieve something? Add "wizards" which force you through the twenty-step process, one slow, painful step at a time. (Converting Xerox Alto's "modeless" paradigm into a good old IBM 704 paradigm...)
People can't figure out how to pop up the "wizards?" Add wizards that pop up by themselves (Clippy).
People get frustrated? Add an automatic frustration detector.
How the heck can you satirize Microsoft when they do such a good job of satirizing themselves?