Mach is a wonderful architecture if you avoid talking about performance. Abstraction = overhead. I saw significant performance increases when I switched xServers from OS X to Suse64.
I'm tired of hearing that excuse. My linux boxes run on an even more diverse set of hw than MS seems to be able to support and I've had a better experience with them.
I'll put it on the list- right after my loved ones master numlock and capitalization. I'm not making any timeline promises though, as my familial help desk still responds to users mired in to-click or double-click conundrums. Or the old quick launch vs. task list quagmire. Outlook isn't the most stable piece of s/w to begin with but it really gets cranky when you open 8 copies of it by double clicking the quick launch instead of single clicking the active windows region of the task bar.
I just a horribly comic vision of a new class of biometric (maybe anthropological) authentication... I won't elaborate but I just know I'm not going to be able to the "hokey pokey" tune out of my head for the rest of the day.
A bit more technical then I was thinking... but I get your drift. I'd love to see an error message, "I'm sorry, this site is not convinced you're human. Could you fill the form out a little slower?"
I was thinking more along the lines "move the mouse in a circle", or "complete the following mouse gesture, L-L-R-U-D" (could even use randomly stylized arrows as a layer of obfuscation)
Agreed. I was trying to convey though that the defensibility is more a function of the number of question generating algorithms, not of any one question.
In response to the people asking about animated gifs, I think they could be algorithmically defeated.
However, what about something requiring mouse movement? For example, using a mouse gesture as an unlocking code. A text (or audio) cue to the user to do something with the mouse.
The above wasn't my first thought after answer the animated gif question. But if follwed from the first thought; instead of animated gifs, what about the Apple Quicktime things that allowed you to move the mouse to view a 3d scene? The entire scene wouldn't be visible and would require mouse movement to view the scene enough to answer the question. Obvious problems- hard to generate. But a mouse gesture based unlocking? Isn't that doable?
I disagree, I think the question domain space (ie. general knowledge, parameteric, etc) is large enough. A few examples have been given in the postings but there are many others and I would think such a system could be maintained over the long run so that as AI+NLP evolves, the questions (or question generating algorithms) could evolve as needed to stay a couple of steps ahead. All the historical & foreseeable future shortcomings of NLP happen to work to this problems advantage. It's kind of a like a shared key concept- as humans we have lots of "shared keys" that bots won't have. Just a couple more families of questions:
1. "Which is T, an x or y?" Where T is some adjective: more expensive, bigger, heavier, smellier, etc.
2. "A is to B as Xn is to Yn" - any sort of general knowledge analogy. Granted these are a bit more IQ Testish but it could be multiple choice with obviously wrong answers.
I don't want to trivialize the cost of maintenance of the "Question Bank", but I think it is a manageable problem.
An obvious concern is to keep your demographics in mind (ie. age group, culture, etc) to make sure there are no prejudicical questions.
That's a *very* nice list. You should write brochures for MS. I certainly won't question the accuracy, but it doesn't change the bottom line- I would get ALOT less done if I had to work in XP. I periodically wonder whether it's a personal perception problem; maybe after many years I'm just tired of investing of myself to learn to maximize my efficiency on yet another version of Windows. I don't think this is it though- I have to make similar investments (granted I have less data due to the shorted time frame) when picking up another x.org paradigm and the payoff seems worthwhile. My gut feeling is this is not the case w/XP. We'll see what happens w/Vista.
My comment had nothing to do with:
1. XP's improved UI over 2k
2. XP's improved ease of management over 2k
3. XP's improved driver support over 2k
I have heard the above from more experienced XP users. I can neither confirm nor deny. What I can say is that the overal performance (CPU+I/O, but excluding 3d) of a 2Ghz/512Mb XP machine is agonizing compared to 1Ghz/256Mb 2K machine when running a standard mix of a typical mix of apps for an IT user.
We all get used to using an environment that "pop's", ie. is responsive given the pace at which an experienced user works. When it doesn't do this, I feel compelled to "float test" the damn thing and go back to a environment where I can get work done (I won't fan the flames by saying which).
Is. Mr. Murray implying that MS has plans to get into the reliable and dependable software business? This could be huge if they're succesful, combined with their existing market penetration this could pose a real threat to the rest of the industry.
I code to the standard and find if I adhere to it I can usually use the same javascript & css for IE + mozilla based browsers with little problems. I do use the validator and find that if everything is valid (HTML, CSS, etc), the site performs more reliably; ie. I spend less time chasing down spurious problems.
You know, for years MarCom has proven that if you have a big enough megaphone, the truth of your words isn't all that relevant. If it's loud enough, it becomes true eventually.
I say we all chip in and get Mr. Dvorak a bigger megaphone.
It must be reassuring to all the professional fud drum bangers that if they ever get tired of IT they always have the option of a succesful career as fantasy and fairy tale authors.
No disrespect to Mr. Dvorak's prior significant contributions to software and computing in general. However, crying the demise of a leviathan like MS is a bit dillusional. Apple went through a period of reinventing themselves succesfully and MS could take decades to do the same if they ever had to. Having a gajillion in cash reserves affords a company that luxury.
Perhaps this move is intended to boost MS's stats at Netcraft. If they convert Yahoo's server farm to MS technology, it's gotta be a significant % increase in the number of MS servers out there. I'd expect that, ironically, this would help Google in the long run.:D
Is it just me or others having ludicrous visuals of a these IT titans embroiled in a celebrity boxing match? Perhaps it's just the Chicken Littles that are embroiled.
It was always my understanding that many of their pieces/parts have a non-MS lineage.
For example: CPM begat MS-DOS, Mosaic begat IE, Sybase begat MS SQL, Hotmail begat... well, MS Hotmail.
I've heard (unconfirmed) that their TCP/IP stack wasn't exactly home grown either.
Slashdot Mirror
Actually, I see this as a great example of software natural selection. The OSS is killing off the weaker software.
Mach is a wonderful architecture if you avoid talking about performance. Abstraction = overhead. I saw significant performance increases when I switched xServers from OS X to Suse64.
I'm tired of hearing that excuse. My linux boxes run on an even more diverse set of hw than MS seems to be able to support and I've had a better experience with them.
Does anybody else find it suspicious that a US vendor's laptop exploded at a Japanese conference? What are the chances that Lenovo was a sponsor?
I'll put it on the list- right after my loved ones master numlock and capitalization. I'm not making any timeline promises though, as my familial help desk still responds to users mired in to-click or double-click conundrums. Or the old quick launch vs. task list quagmire. Outlook isn't the most stable piece of s/w to begin with but it really gets cranky when you open 8 copies of it by double clicking the quick launch instead of single clicking the active windows region of the task bar.
I just a horribly comic vision of a new class of biometric (maybe anthropological) authentication... I won't elaborate but I just know I'm not going to be able to the "hokey pokey" tune out of my head for the rest of the day.
I was thinking more along the lines "move the mouse in a circle", or "complete the following mouse gesture, L-L-R-U-D" (could even use randomly stylized arrows as a layer of obfuscation)
Agreed. I was trying to convey though that the defensibility is more a function of the number of question generating algorithms, not of any one question.
In response to the people asking about animated gifs, I think they could be algorithmically defeated. However, what about something requiring mouse movement? For example, using a mouse gesture as an unlocking code. A text (or audio) cue to the user to do something with the mouse. The above wasn't my first thought after answer the animated gif question. But if follwed from the first thought; instead of animated gifs, what about the Apple Quicktime things that allowed you to move the mouse to view a 3d scene? The entire scene wouldn't be visible and would require mouse movement to view the scene enough to answer the question. Obvious problems- hard to generate. But a mouse gesture based unlocking? Isn't that doable?
I disagree, I think the question domain space (ie. general knowledge, parameteric, etc) is large enough. A few examples have been given in the postings but there are many others and I would think such a system could be maintained over the long run so that as AI+NLP evolves, the questions (or question generating algorithms) could evolve as needed to stay a couple of steps ahead. All the historical & foreseeable future shortcomings of NLP happen to work to this problems advantage. It's kind of a like a shared key concept- as humans we have lots of "shared keys" that bots won't have. Just a couple more families of questions:
1. "Which is T, an x or y?" Where T is some adjective: more expensive, bigger, heavier, smellier, etc.
2. "A is to B as Xn is to Yn" - any sort of general knowledge analogy. Granted these are a bit more IQ Testish but it could be multiple choice with obviously wrong answers.
I don't want to trivialize the cost of maintenance of the "Question Bank", but I think it is a manageable problem.
An obvious concern is to keep your demographics in mind (ie. age group, culture, etc) to make sure there are no prejudicical questions.
That's a *very* nice list. You should write brochures for MS. I certainly won't question the accuracy, but it doesn't change the bottom line- I would get ALOT less done if I had to work in XP. I periodically wonder whether it's a personal perception problem; maybe after many years I'm just tired of investing of myself to learn to maximize my efficiency on yet another version of Windows. I don't think this is it though- I have to make similar investments (granted I have less data due to the shorted time frame) when picking up another x.org paradigm and the payoff seems worthwhile. My gut feeling is this is not the case w/XP. We'll see what happens w/Vista.
I have heard the above from more experienced XP users. I can neither confirm nor deny. What I can say is that the overal performance (CPU+I/O, but excluding 3d) of a 2Ghz/512Mb XP machine is agonizing compared to 1Ghz/256Mb 2K machine when running a standard mix of a typical mix of apps for an IT user.
We all get used to using an environment that "pop's", ie. is responsive given the pace at which an experienced user works. When it doesn't do this, I feel compelled to "float test" the damn thing and go back to a environment where I can get work done (I won't fan the flames by saying which).
If I use XP for more than an hour I feel violently compelled to upgrade- to anything else. Linux or Windows 2000 being perfectly acceptable.
Is. Mr. Murray implying that MS has plans to get into the reliable and dependable software business? This could be huge if they're succesful, combined with their existing market penetration this could pose a real threat to the rest of the industry.
I code to the standard and find if I adhere to it I can usually use the same javascript & css for IE + mozilla based browsers with little problems. I do use the validator and find that if everything is valid (HTML, CSS, etc), the site performs more reliably; ie. I spend less time chasing down spurious problems.
I say we all chip in and get Mr. Dvorak a bigger megaphone.
No disrespect to Mr. Dvorak's prior significant contributions to software and computing in general. However, crying the demise of a leviathan like MS is a bit dillusional. Apple went through a period of reinventing themselves succesfully and MS could take decades to do the same if they ever had to. Having a gajillion in cash reserves affords a company that luxury.
Some sarcasm succeeds, some fails. Sorry.
Stumbling isn't necessarily an obstacle to market domination. I fear the stumblers the most.
Perhaps this move is intended to boost MS's stats at Netcraft. If they convert Yahoo's server farm to MS technology, it's gotta be a significant % increase in the number of MS servers out there. I'd expect that, ironically, this would help Google in the long run. :D
Is it just me or others having ludicrous visuals of a these IT titans embroiled in a celebrity boxing match? Perhaps it's just the Chicken Littles that are embroiled.
Oh yeah, and don't even get me started on MS' Borg like move into the accounting software space (ie. Solomon, Great Plains, etc).
It was always my understanding that many of their pieces/parts have a non-MS lineage. For example: CPM begat MS-DOS, Mosaic begat IE, Sybase begat MS SQL, Hotmail begat... well, MS Hotmail. I've heard (unconfirmed) that their TCP/IP stack wasn't exactly home grown either.
These additional demands are met:
1. a free lollipop.
2. a car ride deep in the forest
Ok, who'll be the first to construct a Lego Beowulf? And I'm sure NTX overclockers have already started!