The Three-Tesla MRI is okay, and clinical use of it is a new development. I've sat in one during its experimental phase and with the wrong pulse sequences, you can actually feel induced currents (or I guess the effects of the induced currents) in your nerves. There are Twelve-Tesla magnets used for imaging mice: one I've seen at CalTech, and quite a few others around the world. The 12-T MRIs can just about see intracellular structures.
The thing they were hyping was using fMRI (functional MRI) which involves making some baseline MRI image acquisitions of the brain (of the BOLD = blood-oxygen level deoxyhemoglobin signal) which depends on the oxygenation state of the hemoglobin molecules in red blood cells. Then, further fMRI images are acquired as the subject performs some mental task. Looking at the difference between these two images shows differences in blood circulation patterns. Initially, there is a drop in the BOLD signal as more oxygen is consumed, then an increase as the local vasculature opens up and more oxygenated hemoglobin flows into the local arteriolar bed. The later increase in BOLD signal is not as specific as the initial decrease in BOLD signal, but gives a general idea of which regions of the brain are activated in correlation with particular mental or physical activities.
This is also often done another way intraoperatively (with the patient awake) by electrically stimulating parts of the brain and seening if a patient's muscles move (if you stimulate the motor cortex) or if the patient senses anything (if you stimulate anteriorly in the sensory cortex). This requires the patient to be awake and is most often done when a patient has a tumor in just deep to the motor cortex or speech areas (Wernicke's / Broca's) in an attempt to avoid injury to "eloquent cortex". This is easier for the patient to deal with if they are adults. I bet pediatric patients wouldn't want to be awakened during surgery and wouldn't be emotionally comfortable with it, so fMRI is probably a cool thing to do for pediatric patients, and useful.
It is NOT, however, novel, as it is being done at almost every neurosurgical center in the US if it is possible and necessary. So their hyping of the fMRI is just hype, but the statements about using the balloon for slowly dissecting open a surgical pathway with pressure is novel and cool.
This is similar to a technique that is used to create more skin for grafting in burn victims: a balloon is implanted underneath the scalp of the pediatric burn victim and is gradually inflated over time. The skin and subdermal tissues are stretched slowly and expand in size, much like the abdomen as we eat too much over the years.
After a month or so, you've got about two-thirds of a sphere of diameter of 8 cm, yielding maybe 128 cm^2 of usable skin for grafting onto the burn victim.
This is a great technique. The trick in surgery is not only taking out what doesn't belong there (the tumor) but leaving intact everything else which does belong there. The slow dissection into the brain teasing apart the structures without damaging them or putting too much pressure on them (which can also damage them by decreasing the blood flow into the area, and hypoxia for greater than a minute can be permanently damaging to neural tissue) or opening up vessels. Creating a tract and then allowing gradual pressure over a long period of time to separate the fascial (I know it's not really fascial, but the equivalent of it) planes seems like a great way to avoid damage. What the article doesn't address is how long a time period this takes place over (as I end this convoluted sentence a preposition with).
The Oklahoma bomber was provably caught the same way: he bought a calling card for long distance calls, but rather than buy new cards as needed, he paid for extra minutes by typing in a credit card number. La voila: linked CC IDentifying data to calls placed, linking his actions and calls and truck rentals.
Rule 1002 of the Federal Rules of Evidence requires that an original document or object be used to prove something. Waiving this rule means that you accept your equipment back, but agree to allowing the FBI's copy of your original hard drive's being deemed as the equivalent of originals.
Rule 1002. Requirement of Original
To prove the content of a writing, recording, or photograph, the original writing, recording, or photograph is required, except as otherwise provided in these rules or by Act of Congress.
I would guess that the smartest thing to do when you receive your evidence back is to make a hash / MD5sum of your enitre hard drive's contents, or individual MD5sums of each partitions' data to corroborate that what you got back matches what you gave. Or is that being too paranoid? Of course, this would work better if your had hashes or off-site backups of your machines that you could also create hashes of. But then if you had off-site backups, this wouldn't have been that much of a problem. Or would the FBI have confiscated the off-site backups too?
I agree with you. I've listened to the agent in San Diego talk about some their information issues at the SDSC regional info watch (sdriw.org) and they seem very professional and with it. If only they could be as responsive to cases that the ISP's bring to their attention.
and the U.S. was trying to place restrictions on the European version of GPS that is being launched that would force the europeans to disable high-accuracy readings if the U.S. wanted them to.
It's similar to the way multiple SBIR contracts are awarded to different companies to enhance competition, find a new way of looking at things, and to get the older more established companies / contractors off their butts. They realize that the gov't doesn't have to keep coming to the established contractors or research universities to get the same things done. But I agree that the leaked route changes the key approach to the concept: responding quickly to a new task without the opportunity for a great amount of preplanning. Connect the dots is a different game. And, in fact, the DARPA Grand Challenge itself was a different game at the start. Until DARPA changed the rules and said that it would limit the challenge to 20 (or so) competitors it would choose. A lot of teams who had funding contingent on being entered officially in the DARPA grand challenge couldn't get their money, and it looked like it was only going to be the "usual suspects" rounding out the contestants list: big time contractors. I have not yet gone to the web site for your group. I also may have a lot of time on my hand. Do you need volunteers and workers? I'd be very interested.
They're not trying to lock up all of this imagery. They're not even trying to charge everyone for using these images. Children doing book-reports, people using these images non-commercially, and others may obtain permission for use of these images just by asking. Go to the top of their RANTS page and read it. They're not being offensively aggressive; they're being defensively aggressive. It looks like they've been burned quite a few times in the past by people trying to hijack their work or even ordering something and cancelling after the majority of the work has been done.
Look at an example of some of their image clean-up work to decide on your own whether they have or have not added some substantial work to images that are in the public domain. The link is
http://cprr.org/Museum/legal.html#Originality
and you can see that the images were not simply cleaned up with a single filter command in Photoshop or GIMP. I've cleaned up some family photographs from the 1900's and 1920's for my mom and considerable work went into cleaning up these negatives and positives. Take a look at those pictures and see if you think they are trying to lock up images from the public domain or trying to make sure that someone else does not profit from their work.
It's not a joke. Look at the text you pointed to: It is very expensive and time consuming to acquire and artistically
restore historic photographs, and we wanted to share thousands of
these wonderful images by placing them online for free viewing by the
public without inadvertently giving away the publication rights which
need to be retained to make our and other museums financially viable. They have in fact done more than simply scan public domain images. They have scanned and cleaned up public domain images: this cleaning up and de-noising is what constitutes the "creative" aspect of their work. They want to retain their copy-rights (see what the word means now?) to be able to reproduce these pictures possibly in book format and sell them to recoup their expenses and perhaps also for profit.
If they didn't have TOS that said DON'T COPY THESE, THESE ARE OURS, someone else could come along and take their photograph clean-up work and profit from them giving none of the benefits of the work to those who produced their work. Now I agree that I can't claim to know how much work they did unless I could also see what the photos' scans looked like before the clean up.
And as for their phone number call charge, it's effectively the way to stop people from spamming their phone number. They're saying: Listen, we're NOT offering licenses to reproduce these images in any other medium to anyone else, DON'T BOTHER calling us to ask for this. IF YOU WASTE MY TIME by calling me at this number, I'll CHARGE YOU.
Isn't this exactly what most./'ers are ranting for on the other stories about unsolicited faxes and unsolicited emails? I may not agree with all of the TOS of that site, but I can at least see the necessity for this from their point of view. They're allowing amateurs to look at these wonderful photos which they've cleaned up, but THEY'RE NOT RELEASING THEM under a BSDish license which would let someone else profit from their work, and they're making it clear that everyone visiting the site is made aware of this.
Intelligent, and a sense of humor, to boot. You must be new here;) That's not the way it's done.
May I ask you whether you constrained the dictionary of probe words before you sent the 10k spams to yourself? Obviously, you attended a conference/meeting at a Marriot (or were watching a lot of Joe Millionaire) for that one word to pop out, or had "white listed" that word for your Bayesian filter. What about doing contextual filtering? Running messages through a parser to check for contextual / grammatical validity would not only be computationally expensive but would also mark many slashdot comments as non-sense; but a parser that checks the immediate predecessor and successor words to see whether a sesame word has just been randomly inserted into text vs. whether it makes sense for that word to be sandwiched as it is.
Yes, it's dedication to research. He sent himself the 10k messages to see if he could outwit his own Bayesian filtering of spam messages. He effectively deduced that if the incoming message can be similar enough to items that have been specifically marked non-spam by the end-user of the Bayesian-spam-filter, it will be not be marked as spam.
There's a cunning recursiveness to this which is at that fine line between clever and stupid. The difficulty is, as he also deduces, that each person's Bayesian rules for spam vs. nonspam are unique and will require many attempt in order to infer the pass-through words that will create a false negative and allow the spam to come through. The one step that people are missing is that if the evil spammer wishes to work on spamming a domain (both in the internet sense and in the "domain of expertise/specialization" sense) she can tailor the pass through words to the market. If she's sending spam to Intel or AMD corporate addresses, then lithography might be the magic word; if she's spamming Xilinx, the fpga will route through the Bayesian filter; if she's spamming Dave Barry, then debenture and fish falling from the sky might help spam make it through, Natalie may or may not make it through a/.'ers filter, actually usually including slashdot in the subject or as the name usually will make it through a slashdotter's filter. And the ease of this lies in that tailoring the open sesame words to a market will probably open the doors to all of the e-mail recipients at a domain, particularly is the spam filtering is done at the mail-server level and not at the end-user level. Thus rather than having to send 10k messages to a single user to crack open the spam doors, sending those 10k messages to multiple users at a domain and analysing which ones get through will effectively open the floodgates for all of the users at that internet domain. And using the concept of a priori probability distributions makes the hunt for these sesame words {[tm]/me:) } easier by limiting the dictionary to be searched to the keywords of the field/domain about to be spammed. That is what makes this dangerous.
The counterattack from the corportate mail-server will be to look for these similarly unique messages being sent to multiple users.
Ask the doctor FOR the tech specs on both of the units along with asking if either unit is a clear cut winner. Ask the doctor how many cochlear devices she has implanted and over how many years; how many of these were in adults and how many were in children. (The FDA approved pediatric implants in the late 1980's; adult implants were being performed earlier.) Ask the ENT doctor what the complication rate has been for HER OPERATIONS at the facility that this implant will be performed. Ask the ENT doctor what the morbidity/mortality is for this procedure. (Cochlear implants are performed, mostly, by ENT specialists, at least in the U.S.) Talk to the doctor yourself to find out what message was actually being conveyed to your mother. Or have your mother return and (since she can't hear) ask for written info for the informed consent which she can take home and review.
More channels equals more recognizability. I've read studies that children with implants have done better in picking up language and speech ability when they have more frequency channels in their implants. I have not seen any studies about adults relating to the number of channels in their cochlear implants. The House Ear Institute in L.A. has also been a site that has done experimental brain-stem auditory implants, interfacing at the cranial-nerve nucleus level rather than at the cochlear nerve.
I agree with your suggestion about the post-surgery support. Find out who the speech therapists and audiologists will be post-operatively for your mother. Talk with them before the operation or before choosing which of the implants should be used.
Don't know that I'd necessarily want it embedded within the seams or the clothing. The wash/dry or dryclean cycles may knock out some of that fancy button power. This may be politically incorrect on/. but even having your own RFID transponder/receiver in one of these would be ool. Instead of labeling the button or the wearer with the RFID tag, your carry-on / wear-along computer gathers the info from the RFID tags around you and then can respond based upon the environment sensed. On my short experience with bluetooth, I'd say that this concept couldn't probably draw too much more battery power than bluetooth.
I wholeheartedly agree. If it's not a problem with the GUI interface (too many gewgaws, unlabeled buttons), it's a problem with menu heirarchies (menu choices in pull down menus with a list length of one, forced step pathways such as having to push the okay button a few times, or of course voice-mail jail menu options), or these days even with the hardware itself. Some of the newer cell phones use admittedly non-standard and amazingly non-utile key-pad layouts or even integrate multiple numbers onto a single rocker switch. But it's like getting used to the problems of Word, Excel, or Explorer: so many people get used to the klunky interfaces that they think it HAS to be that way.
And I definitely agree with your suggestions about talking to people and reading books. And a good walk in the park or next to the cliffs in La Jolla is a great way to recharge the brain cells. Ocean sounds are pleasing too.
When you look at a road partially obscured by raindrops on your windshield, your "visual focus" is doing two things
1 - pointing your eyes to get the correct stereopsis for the object of attention (a.k.a. vergence or crossing your eyes)
2 - focusing (accomodating your lens) the optics to optimize the clarity of the image plane on the object of attention.
Chaning your "visual focus" to the windshield requires changing your vergence to maximize the correlation of the windshield images and changing your accomodation to get that visual plane focussed onto your retina. It's possible to mismatch vergence and accomodation: that is the effect used in the "magic eye" 3-d illusions. The mismatch is also why some people have a hard time appreciating the 3-d effect of "magic eye" pictures, and is also part of the reason for why pilots using HMDs (head mounted displays) get airsick. Part of their brain sees vergence cues telling them that the object they're looking at is 20 meters away, but the focus cues (and parallax cues and a few other things, include visual lag time) tell them that the focal plane is only a few centimeters away.
To get a 3-d effect of being able to change your "visual focus", you need to be able to track your eye position AND you need to be able to present a true 3-dimensional display or at least separate visual displays for each of the eyes which you can re-merge with your brain into your "3-d" image of the world.
These "3-d" cards simply hardware accelerate the rendering of the 2-d projection of 3-d data and textures. The effect you'd like would require the presence of multiple planes of imaging to focus on. There actually have been a couple of LCD dispays that use multiple display screens at different physical depths to give a partial 3-dimensional visual effect. These, however, differ so slightly in distance that the vergence/accomodation effects on the eye might not be very significant.
And he's also been working on the concepts of consciousness and intelligence and memory alongside Christof Koch at CalTech. He's published a book recently on some of this.
Your example gave the opposite result for me. The 'Q' actually has a descender on it and "pops out" visually from the line with Os on it, whereas the 'T' is confined to the same vertical limits as the Os and does not visually "pop out." This is obviously also a function of the font being used and not a disagreement with your premise. I'm trying to think of the link on visual crowding and visual psychophysics for this.
The other thing that is important in symbology (which is important in designing air traffic controller displays, or military displays with icons or symbols or letters representing individual troops or armaments or hardware) along with recognizability is differentiability. If you've got multiple symbols on a crowded display, how do you make particular features stand out and how do you make multidimensional representations in symbols and icons?
One way is to have the symbol represent one thing (type of object) and its color represent another thing (category of object, e.g. increasing altitude, decreasing altitude, steady altitude, or red and green for the opposing positions in a football team instead of Xs and Os).
And small symbols subtend less than half-a-degree of visual arc, they are extremely tiny. Most of the color matching and color discrimination work has been done using 10 degree and 2 degree spots. The CIE XYZ and CIELAB color spaces (from the 1930s and the 1970s) all used relatively large targets compared to the size of icons on the original Mac or text on current screens.
I've got a PowerMac 8100 that occupies a position of power in the garage with its 21 inch Apple monitor on a heavy duty shelf. I tend to start it and walk back into the kitchen during its approximately 2 minute long bootup sequence.
The 8100 bootup crash sound is a sound sample of brakes screeching followed by a loud crash of vehicle into something immobile. I ran into the garage when I heard that sound a year ago thinking that a car had crashed through my garage door. Yikes! It was just the CMOS battery that had run low which triggered it, and I rebooted it and figured out that it was the Mac that crashed, replaced the battery a week later and I haven't heard that sound since.
Now that's an easter egg. BTW, if you have an old Mac 512k, take it apart and look on the inside of the shell casing: you'll find the signatures of the Mac development team in it.
So there we have it. A triple duplicate posting to fool some of the./ers some of the time, especially those who do not look at the top page simultaneously containing all three postings at once! This was obviously done to counteract the comment in the first posting stating that if this were a REAL story, there'd be duplicate submissions.
But really folks, a duplicate submission by the same editor within 148 minutes? Outrageous!!!
New game for those tired of google-games: find the shortest interval duplicate submission. Easy attack mode: look for duplicated links, e.g. ftp://ftp.rfc-editor.org/in-notes/rfc3514.txt, as the giveaway and compare posting times.
New idea for a submission filter for the editors: implement the comparison of links concept to weed out duplicate postings by editors. If this were really implemented in the slashcode, however, it could destroy the whole nature of reality for./ers who have come to expect, nay rely upon, the duplicates.
I didn't use the word "dupe" because I must admit the first time I saw an editorial comment saying "yes it's a dupe", I thought someone had been duped into believing something that is untrue. Of course, that's an everyday occurence here, not something which is reserved for April 1st.
I must admit that I enjoy these April Fools' items.
I've seen similar errors made by vision science (note that I did not say "image processing") researchers trying to analyze natural scene statistics and come up with interesting patterns. They created "basis functions" and did principal component analysis on sets of images and came up with a basis set that looks curiously like the base images of the DCT (discrete cosine transform), the underlying calculations of the JPEG image format. This is to be expected when you start with a set of images that are JPEG compressed.
This was actually published in a (barely) peer-reviewed journal, Vision Research. I didn't say "image processing" above because a lot of these vision scientists seem to be psycologists doing visual psychophysics without having a strong background in math, or optics, or (it seems at time) the fundamentals of science.
The other thing to take into consideration is that gzip is "pseudolinear". It does not take into account the 2-dimensional correlations that exist in image data. Even fax compression takes advantage of it. (and yes, I do realize that gzip can account for runs from previous regions regardless of length or location, but I am trying to point out that there is a specific 2-dimensional set of correlations extant in 2-d image data).
In these cases being cited that use GZIP, the major function of GZIP seems to be as an indicator of the presence or absence of high-frequency components in the signal stream. Lots of irregular high frequency -> Low compressibility, very little irregular high frequency --> High compressibility factors.
Try ZipSlack, a It's a ~37MB download which can be installed into a DOS partition and expanded there and started from DOS mode.
Believe in it: I put it onto a 120-MB laptop hard-drive and got it to run its 2.4.18 kernel on my amazingly non-decked-out AST PowerExec Laptop which has a 386/25 processor in it. Yes, the timer comes out at 6.something BOGOMIPS and it runs slowly, but it runs and FITS on a small piece of hardware.
I got the ZIPped version of the file onto the laptop HD by using an IDE-to-laptop adaptor cable, and used the massive power (relatively) of an AMD-K6-333 processor to unzip the file before reinserting the HD back into the laptop.
I'm still tinkering with it and need to remove some of the ISA/PCI packages, etc, and try to get the PCMIA card to work with it, but the command line tools work, and I was able to run some key console mode apps I compiled (on the AMD) on the laptop without any problem.
In both of these cases, calling the 1-800-xxx-yyyy number created an account entry with the phone numbe you were calling from on it. This account entry was provided to the owner of the 1-800-number, who used it to call you back.
While a 1-{800 | 888 | 877 | 866} number is free to the calling party, except for some nefarious call redirection scams, it is NOT free to the receiving party. They pay for the call. They can receive ANI information detailing which phone number is calling them as long as they pay extra for it.
Unlike Caller-ID information which is transmitted in-band (on the same line) between the first and second telephone ring and can be blocked by the dialing party, the ANI service is transmitted off-band and CANNOT be blocked (either one time or permanently) when you call an 800 number. It's always there.
Slashdot Mirror
The Three-Tesla MRI is okay, and clinical use of it is a new development. I've sat in one during its experimental phase and with the wrong pulse sequences, you can actually feel induced currents (or I guess the effects of the induced currents) in your nerves. There are Twelve-Tesla magnets used for imaging mice: one I've seen at CalTech, and quite a few others around the world. The 12-T MRIs can just about see intracellular structures.
The thing they were hyping was using fMRI (functional MRI) which involves making some baseline MRI image acquisitions of the brain (of the BOLD = blood-oxygen level deoxyhemoglobin signal) which depends on the oxygenation state of the hemoglobin molecules in red blood cells. Then, further fMRI images are acquired as the subject performs some mental task. Looking at the difference between these two images shows differences in blood circulation patterns. Initially, there is a drop in the BOLD signal as more oxygen is consumed, then an increase as the local vasculature opens up and more oxygenated hemoglobin flows into the local arteriolar bed. The later increase in BOLD signal is not as specific as the initial decrease in BOLD signal, but gives a general idea of which regions of the brain are activated in correlation with particular mental or physical activities.
This is also often done another way intraoperatively (with the patient awake) by electrically stimulating parts of the brain and seening if a patient's muscles move (if you stimulate the motor cortex) or if the patient senses anything (if you stimulate anteriorly in the sensory cortex). This requires the patient to be awake and is most often done when a patient has a tumor in just deep to the motor cortex or speech areas (Wernicke's / Broca's) in an attempt to avoid injury to "eloquent cortex". This is easier for the patient to deal with if they are adults. I bet pediatric patients wouldn't want to be awakened during surgery and wouldn't be emotionally comfortable with it, so fMRI is probably a cool thing to do for pediatric patients, and useful.
It is NOT, however, novel, as it is being done at almost every neurosurgical center in the US if it is possible and necessary. So their hyping of the fMRI is just hype, but the statements about using the balloon for slowly dissecting open a surgical pathway with pressure is novel and cool.
This is similar to a technique that is used to create more skin for grafting in burn victims: a balloon is implanted underneath the scalp of the pediatric burn victim and is gradually inflated over time. The skin and subdermal tissues are stretched slowly and expand in size, much like the abdomen as we eat too much over the years.
After a month or so, you've got about two-thirds of a sphere of diameter of 8 cm, yielding maybe 128 cm^2 of usable skin for grafting onto the burn victim.
This is a great technique. The trick in surgery is not only taking out what doesn't belong there (the tumor) but leaving intact everything else which does belong there. The slow dissection into the brain teasing apart the structures without damaging them or putting too much pressure on them (which can also damage them by decreasing the blood flow into the area, and hypoxia for greater than a minute can be permanently damaging to neural tissue) or opening up vessels. Creating a tract and then allowing gradual pressure over a long period of time to separate the fascial (I know it's not really fascial, but the equivalent of it) planes seems like a great way to avoid damage. What the article doesn't address is how long a time period this takes place over (as I end this convoluted sentence a preposition with).
Ha ha ha
Just saw some of the beta versions of One Note,
including the Tablet computer version of One Note
on a compaq tablet at the UCSD bookstore.
Try doing anything on it and it goes to a dialog box with "Sorry, your beta version of One Note is expired software." Now That's How to Sell (TM) !
The Oklahoma bomber was provably caught the same way: he bought a calling card for long distance calls, but rather than buy new cards as needed, he paid for extra minutes by typing in a credit card number. La voila: linked CC IDentifying data to calls placed, linking his actions and calls and truck rentals.
Rule 1002 of the Federal Rules of Evidence requires that an original document or object be used to prove something. Waiving this rule means that you accept your equipment back, but agree to allowing the FBI's copy of your original hard drive's being deemed as the equivalent of originals.
Rule 1002. Requirement of Original
To prove the content of a writing, recording, or photograph, the original writing, recording, or photograph is required, except as otherwise provided in these rules or by Act of Congress.
I would guess that the smartest thing to do when you receive your evidence back is to make a hash / MD5sum of your enitre hard drive's contents, or individual MD5sums of each partitions' data to corroborate that what you got back matches what you gave. Or is that being too paranoid? Of course, this would work better if your had hashes or off-site backups of your machines that you could also create hashes of. But then if you had off-site backups, this wouldn't have been that much of a problem. Or would the FBI have confiscated the off-site backups too?
I agree with you. I've listened to the agent in San Diego talk about some their information issues at the SDSC regional info watch (sdriw.org) and they seem very professional and with it. If only they could be as responsive to cases that the ISP's bring to their attention.
and the U.S. was trying to place restrictions on the European version of GPS that is being launched that would force the europeans to disable high-accuracy readings if the U.S. wanted them to.
l ileo/doc/gal_european_dependence_on_gps_rev22.pdf
The european system is called Galileo. pdf link to http://europa.eu.int/comm/dgs/energy_transport/ga
It's similar to the way multiple SBIR contracts are awarded to different companies to enhance competition, find a new way of looking at things, and to get the older more established companies / contractors off their butts.
They realize that the gov't doesn't have to keep coming to the established contractors or research universities to get the same things done.
But I agree that the leaked route changes the key approach to the concept: responding quickly to a new task without the opportunity for a great amount of preplanning. Connect the dots is a different game. And, in fact, the DARPA Grand Challenge itself was a different game at the start.
Until DARPA changed the rules and said that it would limit the challenge to 20 (or so) competitors it would choose. A lot of teams who had funding contingent on being entered officially in the DARPA grand challenge couldn't get their money, and it looked like it was only going to be the "usual suspects" rounding out the contestants list: big time contractors.
I have not yet gone to the web site for your group. I also may have a lot of time on my hand. Do you need volunteers and workers? I'd be very interested.
They're not trying to lock up all of this imagery. They're not even trying to charge everyone for using these images. Children doing book-reports, people using these images non-commercially, and others may obtain permission for use of these images just by asking. Go to the top of their RANTS page and read it. They're not being offensively aggressive; they're being defensively aggressive. It looks like they've been burned quite a few times in the past by people trying to hijack their work or even ordering something and cancelling after the majority of the work has been done.
Look at an example of some of their image clean-up work to decide on your own whether they have or have not added some substantial work to images that are in the public domain. The link is
http://cprr.org/Museum/legal.html#Originality
and you can see that the images were not simply cleaned up with a single filter command in Photoshop or GIMP. I've cleaned up some family photographs from the 1900's and 1920's for my mom and considerable work went into cleaning up these negatives and positives. Take a look at those pictures and see if you think they are trying to lock up images from the public domain or trying to make sure that someone else does not profit from their work.
It's not a joke. Look at the text you pointed to: It is very expensive and time consuming to acquire and artistically ./'ers are ranting for on the other stories about unsolicited faxes and unsolicited emails? I may not agree with all of the TOS of that site, but I can at least see the necessity for this from their point of view. They're allowing amateurs to look at these wonderful photos which they've cleaned up, but THEY'RE NOT RELEASING THEM under a BSDish license which would let someone else profit from their work, and they're making it clear that everyone visiting the site is made aware of this.
restore historic photographs, and we wanted to share thousands of
these wonderful images by placing them online for free viewing by the
public without inadvertently giving away the publication rights which
need to be retained to make our and other museums financially viable.
They have in fact done more than simply scan public domain images. They have scanned and cleaned up public domain images: this cleaning up and de-noising is what constitutes the "creative" aspect of their work. They want to retain their copy-rights (see what the word means now?) to be able to reproduce these pictures possibly in book format and sell them to recoup their expenses and perhaps also for profit.
If they didn't have TOS that said DON'T COPY THESE, THESE ARE OURS, someone else could come along and take their photograph clean-up work and profit from them giving none of the benefits of the work to those who produced their work. Now I agree that I can't claim to know how much work they did unless I could also see what the photos' scans looked like before the clean up.
And as for their phone number call charge, it's effectively the way to stop people from spamming their phone number. They're saying: Listen, we're NOT offering licenses to reproduce these images in any other medium to anyone else, DON'T BOTHER calling us to ask for this. IF YOU WASTE MY TIME by calling me at this number, I'll CHARGE YOU.
Isn't this exactly what most
``particularly is the spam filtering is done at the mail-server level''
is a typo (and I previewed too). What I meant to type is:
particularly if the spam filtering is done at the mail-server level
Intelligent, and a sense of humor, to boot. You must be new here ;) That's not the way it's done.
May I ask you whether you constrained the dictionary of probe words before you sent the 10k spams to yourself? Obviously, you attended a conference/meeting at a Marriot (or were watching a lot of Joe Millionaire) for that one word to pop out, or had "white listed" that word for your Bayesian filter. What about doing contextual filtering? Running messages through a parser to check for contextual / grammatical validity would not only be computationally expensive but would also mark many slashdot comments as non-sense; but a parser that checks the immediate predecessor and successor words to see whether a sesame word has just been randomly inserted into text vs. whether it makes sense for that word to be sandwiched as it is.
Yes, it's dedication to research. He sent himself the 10k messages to see if he could outwit his own Bayesian filtering of spam messages. He effectively deduced that if the incoming message can be similar enough to items that have been specifically marked non-spam by the end-user of the Bayesian-spam-filter, it will be not be marked as spam.
/.'ers filter, actually usually including slashdot in the subject or as the name usually will make it through a slashdotter's filter. And the ease of this lies in that tailoring the open sesame words to a market will probably open the doors to all of the e-mail recipients at a domain, particularly is the spam filtering is done at the mail-server level and not at the end-user level. Thus rather than having to send 10k messages to a single user to crack open the spam doors, sending those 10k messages to multiple users at a domain and analysing which ones get through will effectively open the floodgates for all of the users at that internet domain. And using the concept of a priori probability distributions makes the hunt for these sesame words {[tm] /me :) } easier by limiting the dictionary to be searched to the keywords of the field/domain about to be spammed. That is what makes this dangerous.
There's a cunning recursiveness to this which is at that fine line between clever and stupid. The difficulty is, as he also deduces, that each person's Bayesian rules for spam vs. nonspam are unique and will require many attempt in order to infer the pass-through words that will create a false negative and allow the spam to come through. The one step that people are missing is that if the evil spammer wishes to work on spamming a domain (both in the internet sense and in the "domain of expertise/specialization" sense) she can tailor the pass through words to the market. If she's sending spam to Intel or AMD corporate addresses, then lithography might be the magic word; if she's spamming Xilinx, the fpga will route through the Bayesian filter; if she's spamming Dave Barry, then debenture and fish falling from the sky might help spam make it through, Natalie may or may not make it through a
The counterattack from the corportate mail-server will be to look for these similarly unique messages being sent to multiple users.
Ask the doctor FOR the tech specs on both of the units along with asking if either unit is a clear cut winner. Ask the doctor how many cochlear devices she has implanted and over how many years; how many of these were in adults and how many were in children. (The FDA approved pediatric implants in the late 1980's; adult implants were being performed earlier.) Ask the ENT doctor what the complication rate has been for HER OPERATIONS at the facility that this implant will be performed. Ask the ENT doctor what the morbidity/mortality is for this procedure. (Cochlear implants are performed, mostly, by ENT specialists, at least in the U.S.) Talk to the doctor yourself to find out what message was actually being conveyed to your mother. Or have your mother return and (since she can't hear) ask for written info for the informed consent which she can take home and review.
More channels equals more recognizability. I've read studies that children with implants have done better in picking up language and speech ability when they have more frequency channels in their implants. I have not seen any studies about adults relating to the number of channels in their cochlear implants. The House Ear Institute in L.A. has also been a site that has done experimental brain-stem auditory implants, interfacing at the cranial-nerve nucleus level rather than at the cochlear nerve.
I agree with your suggestion about the post-surgery support. Find out who the speech therapists and audiologists will be post-operatively for your mother. Talk with them before the operation or before choosing which of the implants should be used.
Don't know that I'd necessarily want it embedded within the seams or the clothing. The wash/dry or dryclean cycles may knock out some of that fancy button power. This may be politically incorrect on /. but even having your own RFID transponder/receiver in one of these would be ool. Instead of labeling the button or the wearer with the RFID tag, your carry-on / wear-along computer gathers the info from the RFID tags around you and then can respond based upon the environment sensed. On my short experience with bluetooth, I'd say that this concept couldn't probably draw too much more battery power than bluetooth.
I wholeheartedly agree. If it's not a problem with the GUI interface (too many gewgaws, unlabeled buttons), it's a problem with menu heirarchies (menu choices in pull down menus with a list length of one, forced step pathways such as having to push the okay button a few times, or of course voice-mail jail menu options), or these days even with the hardware itself. Some of the newer cell phones use admittedly non-standard and amazingly non-utile key-pad layouts or even integrate multiple numbers onto a single rocker switch.
But it's like getting used to the problems of Word, Excel, or Explorer: so many people get used to the klunky interfaces that they think it HAS to be that way.
And I definitely agree with your suggestions about talking to people and reading books. And a good walk in the park or next to the cliffs in La Jolla is a great way to recharge the brain cells. Ocean sounds are pleasing too.
Yeah, I was getting tired of that paperboy following me down those ski slopes and recreating Hitchcock-esque scenes when I tried to get into my car.
When you look at a road partially obscured by raindrops on your windshield, your "visual focus" is doing two things
1 - pointing your eyes to get the correct stereopsis for the object of attention (a.k.a. vergence or crossing your eyes)
2 - focusing (accomodating your lens) the optics to optimize the clarity of the image plane on the object of attention.
Chaning your "visual focus" to the windshield requires changing your vergence to maximize the correlation of the windshield images and changing your accomodation to get that visual plane focussed onto your retina. It's possible to mismatch vergence and accomodation: that is the effect used in the "magic eye" 3-d illusions. The mismatch is also why some people have a hard time appreciating the 3-d effect of "magic eye" pictures, and is also part of the reason for why pilots using HMDs (head mounted displays) get airsick. Part of their brain sees vergence cues telling them that the object they're looking at is 20 meters away, but the focus cues (and parallax cues and a few other things, include visual lag time) tell them that the focal plane is only a few centimeters away.
To get a 3-d effect of being able to change your "visual focus", you need to be able to track your eye position AND you need to be able to present a true 3-dimensional display or at least separate visual displays for each of the eyes which you can re-merge with your brain into your "3-d" image of the world.
These "3-d" cards simply hardware accelerate the rendering of the 2-d projection of 3-d data and textures. The effect you'd like would require the presence of multiple planes of imaging to focus on. There actually have been a couple of LCD dispays that use multiple display screens at different physical depths to give a partial 3-dimensional visual effect. These, however, differ so slightly in distance that the vergence/accomodation effects on the eye might not be very significant.
And he's also been working on the concepts of consciousness and intelligence and memory alongside Christof Koch at CalTech. He's published a book recently on some of this.
Your example gave the opposite result for me. The 'Q' actually has a descender on it and "pops out" visually from the line with Os on it, whereas the 'T' is confined to the same vertical limits as the Os and does not visually "pop out." This is obviously also a function of the font being used and not a disagreement with your premise. I'm trying to think of the link on visual crowding and visual psychophysics for this.
The other thing that is important in symbology (which is important in designing air traffic controller displays, or military displays with icons or symbols or letters representing individual troops or armaments or hardware) along with recognizability is differentiability. If you've got multiple symbols on a crowded display, how do you make particular features stand out and how do you make multidimensional representations in symbols and icons?
One way is to have the symbol represent one thing (type of object) and its color represent another thing (category of object, e.g. increasing altitude, decreasing altitude, steady altitude, or red and green for the opposing positions in a football team instead of Xs and Os).
And small symbols subtend less than half-a-degree of visual arc, they are extremely tiny. Most of the color matching and color discrimination work has been done using 10 degree and 2 degree spots. The CIE XYZ and CIELAB color spaces (from the 1930s and the 1970s) all used relatively large targets compared to the size of icons on the original Mac or text on current screens.
I've got a PowerMac 8100 that occupies a position of power in the garage with its 21 inch Apple monitor on a heavy duty shelf. I tend to start it and walk back into the kitchen during its approximately 2 minute long bootup sequence.
The 8100 bootup crash sound is a sound sample of brakes screeching followed by a loud crash of vehicle into something immobile. I ran into the garage when I heard that sound a year ago thinking that a car had crashed through my garage door. Yikes! It was just the CMOS battery that had run low which triggered it, and I rebooted it and figured out that it was the Mac that crashed, replaced the battery a week later and I haven't heard that sound since.
Now that's an easter egg. BTW, if you have an old Mac 512k, take it apart and look on the inside of the shell casing: you'll find the signatures of the Mac development team in it.
But really folks, a duplicate submission by the same editor within 148 minutes? Outrageous!!!
New game for those tired of google-games: find the shortest interval duplicate submission. Easy attack mode: look for duplicated links, e.g. ftp://ftp.rfc-editor.org/in-notes/rfc3514.txt, as the giveaway and compare posting times.
New idea for a submission filter for the editors: implement the comparison of links concept to weed out duplicate postings by editors. If this were really implemented in the slashcode, however, it could destroy the whole nature of reality for ./ers who have come to expect, nay rely upon, the duplicates.
I didn't use the word "dupe" because I must admit the first time I saw an editorial comment saying "yes it's a dupe", I thought someone had been duped into believing something that is untrue. Of course, that's an everyday occurence here, not something which is reserved for April 1st.
I must admit that I enjoy these April Fools' items.
This was actually published in a (barely) peer-reviewed journal, Vision Research. I didn't say "image processing" above because a lot of these vision scientists seem to be psycologists doing visual psychophysics without having a strong background in math, or optics, or (it seems at time) the fundamentals of science.
The other thing to take into consideration is that gzip is "pseudolinear". It does not take into account the 2-dimensional correlations that exist in image data. Even fax compression takes advantage of it. (and yes, I do realize that gzip can account for runs from previous regions regardless of length or location, but I am trying to point out that there is a specific 2-dimensional set of correlations extant in 2-d image data).
In these cases being cited that use GZIP, the major function of GZIP seems to be as an indicator of the presence or absence of high-frequency components in the signal stream. Lots of irregular high frequency -> Low compressibility, very little irregular high frequency --> High compressibility factors.
Believe in it: I put it onto a 120-MB laptop hard-drive and got it to run its 2.4.18 kernel on my amazingly non-decked-out AST PowerExec Laptop which has a 386/25 processor in it. Yes, the timer comes out at 6.something BOGOMIPS and it runs slowly, but it runs and FITS on a small piece of hardware.
I got the ZIPped version of the file onto the laptop HD by using an IDE-to-laptop adaptor cable, and used the massive power (relatively) of an AMD-K6-333 processor to unzip the file before reinserting the HD back into the laptop.
I'm still tinkering with it and need to remove some of the ISA/PCI packages, etc, and try to get the PCMIA card to work with it, but the command line tools work, and I was able to run some key console mode apps I compiled (on the AMD) on the laptop without any problem.
While a 1-{800 | 888 | 877 | 866} number is free to the calling party, except for some nefarious call redirection scams, it is NOT free to the receiving party. They pay for the call. They can receive ANI information detailing which phone number is calling them as long as they pay extra for it.
Unlike Caller-ID information which is transmitted in-band (on the same line) between the first and second telephone ring and can be blocked by the dialing party, the ANI service is transmitted off-band and CANNOT be blocked (either one time or permanently) when you call an 800 number. It's always there.