Black Hats vs White Hats: Why is it relevent to the issue? How is it measureable or documented?
You're quite right, it is not easily measured, but it is widely accepted that security holes are often discovered through the act of careless exploits.
About the number of security holes: No one can know about security holes that 'no one' knows about. This is true of all OS/webserver combos. I guess it's relevent that M$ isn't disclosing it's source-but that only means that we cannot fix holes we find.
It is infinately more difficult to reverse engineer a product than it is to look at the source and study it for weaknesses. At the very least, the source code acts as a guide to explore potential vulnerabilities.
As fer incentive: Apache provides no incentive to investigate the holes. It is only the case that hackers, white or black, tend to investigate holes for their own reasons, independent of the vendor. NASDAQ is a big enough site that people will try to hack it even if it's running an Open Source package.
While both IIS and Apache provide people with ample kudos for finding security holes, the attitudes are different. You can't even own a copy of IIS without shelling out for NT server, and then when you do, reverse-engineering puts you in violation of your license agreement. If you were to approach MS with a hole, and somehow convince them that it is a serious issue, you'll be lucky if you're not arrested. If not for piracy, for violation of your License... or you could report it, just give MS a short time to act on the bug, exploit it, make a name for yourself in the news and maybe let a few tools slip.
Hidden developers, lack of source, and potential legal consequences are all disincentive. The only reason to do them the favour when you just spent weeks hacking through a bug, is in fear of their applications failing.
Apache is so much easier. Just post the bug to the developers and be laughed at or be thanked. It's like debugging code written by your own company.
Open Source projects doesn't inhibit people from *fixing* security holes. Finding the hole is as easy as exploiting it, and people are always trying to find holes to exploit.
Finding the hole is nowhere near as easy as exploiting it. Not having the source is a major inhibitor to studying the security of an application. Reverse-engineering bugs is a pain in the butt...
Does anybody see an IM application using Gnutella like protocols? Register yourself with your friends and maybe one of thousands of semi-static servers and totally negate the need for AOL/Netscape/Mirabilis and... who's left... Microsoft and Yahoo/Geocities?
I get tired of people crying hypocracy when an open forum appears to draw hippcritial conclusions. These are not the opinions of individuals. They are the opinions of those within a group who cry the loudest.
Some people scream about how much the RIAA sucks and how copyrighted music should be free, others scream about the GPL and how it is being violated. There is no reason to believe that these are the same people.
Personally I love the GPL, it lets me write derivative software at work without losing my soul. I hate the RIAA, because I can no longer bear to listen to the radio, television or CDs.
Honour copyrights, not the people who use them to exploit others.
Pirating bad music taken from exploited musicians doesn't help anybody.
This is a very simple idea which hit me when I was walking down the street one day.
All the crappy musicians we here about these days are only heard about because the record industry promotes them. If there were an alternative channel to promote music, maybe I could begin to find some music which doesn't make me wretch. I mean I don't have cable, and I can't bear to listen to the radio... all the loud carpet-shop commercials, "zoom-zoom-zoom" tv-like ads, repeated songs and manipulated news... it just sucks.
What if somebody were to write a magazine which reviews small musicians, and sponsors them to produce MP3s. With a large enough startup capital, they could pay for producers and the like.
The magazine could have two sections... one for musicians which showed remarkable talent and were sponsored. Another section for musicians who are willing to pay to promote themselves. Self-promoting musicians would be clearly indicated... a condition of this could be that the musician has an MP3 file available, and that the musician themselves own their own name and music.
So you sit back, read your magazine, download the MP3s of some bands you've read about etcetera. Maybe it could form a niche. Maybe you don't need radio stations and record labels to popularize music... you could use night clubs, live bands, print and MP3s
This way, the slew of crappy MP3s are filtered out, people have to read to get a sense of which musicians are worthwhile... musicians get some publicity and could potentially earn cash for selling downloads of MP3s...
You could scrap the meatspace printed matter when flat panels become as durable, cheap, high resolution, low power consumption and establish as wide a circulation.
Then it occurred to me that if I've thought of it, somebody else is probably already doing it.
DeCSS wouldn't have come about so easily if the author didn't study the encryption mechanisms in Xing's player. That's not nothing.
Otherwise, I agree. It sounds like the ability to destroy another company for using some fundamental concept (such as scrolling left to right, using cascading menus, shooting falling objects from a stationary target, editing text with a keyboard, arrow keys and pgup/pgdn and soforth) is directly proportional to the size of their legal department.
These Hasbro clones on the other hand used the Hasbro games and the strong similarities of the name of the Hasbro games to market their products. If Tetris were called "Blocks" I could understand the re-use. Microsoft certainly couldn't sue for people using the word "Word" in the name of their wordprocessor (which is MS Word compatible, and edits text...with a keyboard, menus, dictionary, etcetera).
There has to be a line somewhere... There is a company which owns the use of the word "Ethical" in the context of Mutual funds. Whenever a company creates an "Ethical" fund and calls it "Ethical" the legal department is engaged. On the other hand, their own funds need not necessarily be "Ethical".
I suppose my point is that it is just as dumb and confused outside the computer industry as it is inside.
The "What did you do wrong?" attitude is the same defense used by people who claim that they've been running Win'95 for months without rebooting.
I was installing off a Sony CDU31a on an RLL HDD. That's an XT drive with a proprietary CDROM. Digging up the correct kernel, finding boot parameters, and trying to figure out why I was getting CDR101 read errors while installing took a month.
Come to think of it, I never did figure out the read errors. The drive worked flawlessly under DOS, Win3.1 and OS/2. I wound up copying packages to my FAT partition to install the base packages. Any time I had to read more than a few megs, the CDROM would throw me errors. Despite all this, it was great for installing individual packages after the base was installed.
With an IDE HDD and IDE CDROM, it might have taken me 5 hours...
It took me roughly a month to install Slackware back in 1995. I didn't have the internet to help me, and I had some pretty unusual hardware, but DOS and Windows didn't have any trouble at all.
Windows 3.1 was by far more memory efficient and faster. X was unusable on an unaccellerated 640x480 ISA VGA adapter... unlike Win3.1.
I did some Fortran programming under the environment, which I found useful... but it was easier to dial into the VAX. The fonts were terrible, printer support was almost nonexistant, so for me it was effectively a bloated programming environment.
I administered a small dial up ISP for a short period in 1997. They were running NT3.51, and were scrambling to try to put together a decent system for software development... I put the whole operation under Linux. No big deal. Everything was completely free and worked flawlessly.
Since 1995, I've tried many times to establish a reasonable user environment under Linux. I've tried Gnome and KDE, and they haven't done anything other than to promise me what I had on a 386 running Win3.1... that is to say a reasonable printing architecture, some decent fonts, some standard keystrokes and cut-and-paste. I've tried simplifying things... for some time I ran ICEWM (which I really do like) and then proceeded to try to set up an email client.
There isn't anything but promises here too. Netscape appears to be the best choice... unless I go back to PINE. So I tried to configure Netscape... it crashes. I try some more, it crashes again... infact it crashes quite reliably under many different scenarios... none of which are preventable. So I give up. I tried Spruce, which though promising, was simply incomplete.
I then try to hear multiple audio streams simultaneously. After installing ALSA, reading many FAQs and getting it running, I learn that the Advanced Linux Sound Architecture doesn't support multiple simultaneous streams... like Win3.1 did. So I installed ESD, then I installed Real Player. After a week of applying various binary patches, reinstalling, reading reams of documentation describing how Real Inc. depended upon undocumented features and broke in V5, I finally got some alpha version of the player working and started listening to some Internet radio broadcasts.
So I began working on tracking down another email client. Inches from headed back to PINE, reading up on fetchmail and procmail, Real Player freezes.
Many other experiments revealed that it always would do this some 15 minutes into the broadcast.
On another forray into the Linux world, I installed Gnomehack... it worked fine for a while, and I reminisced about playing hack for long hours... When I went to restore a game one day, X crashed.
It was remarkable. I launch the game, Gnomehack and X crashes, taking down all my other applications, loosing all my data, reliably, reproducably, and inexplicably. It obviously is a bug in my X server.
After many forrays, and certainly many more to come, I have come to the conclusion that the best server OS for a small company or even for a home user is Linux. It is stable, not price prohibative, will teach you a lot about how computers work, and has many other bennefits. The best GUI for that operating system is either Windows, MacOS, OS/2 or just about anything else you can think of. Linux handles itself so well in a networked environment that there is no reason to put a keyboard or monitor on the device.
I can use Pine just as well on my remote linux box as I can locally, and I'll even be able to fill the fonts crisply out to the edge of the screen, and even cut and paste from it.... while listening to radio broadcasts, and chatting on that whore of a program called ICQ... and I will spend less time reinstalling the operating system for the life of the hardware than I will configuring Linux to do the same.
To Linux's credit, Linux is closer to a good UI than Microsoft is to a robust multiuser operating system.
Apply for a teaching position at a university? Maybe they'll print your book... yes, they may charge in excess of $200 for it, but you'll be paid to teach the material you're writing about.
Personally I don't see the value of printing the book. I imagine O'Reilly isn't interested because it wasn't a good seller. This is no reflection on the quality of the book (I've skimmed it and put it back on the shelf), just a reflection on the size of the market.
(For those going to respond that Linux is huge, yadda yadda yadda, I mean a small market of Linux developers which work on TCP/IP programs, and have found the existing code, documentation and community support insufficient.)
The market of people willing to sign up for a detailed course on Network protocols, at which the core material is Linux's TCP/IP stack might be much more lucrative.
For all you know he might be writing on a machine which doesn't have lowercase characters... like... well... god, I have no idea what you would use to connect from your office/whatever in those days...
A Timex Sinclair 1000 desparately soldered togther to emulate an RS232 port?
Unless there were actually VTs which were case impaired....
Maybe a PET which would require switching video modes to see 40 columns of lower case characters.
Because it won't be a CD, it will be an audio-DVD. Also, the DMCA makes any attempts to circumvent the system illegal -- even going down to Radio Shack and buying an Op-amp so that you can make a backup copy of the media or use soundbites from the disk for your default beep.
They're not the best for portability or battery life, but they would be perfect candidates for what the Cursoe chip claims to be. Have you ever seen a 240?
I have to wonder if it will even help with retina detachments, the nerve endings wouldn't be there to interface with... it seems to be limited to helping people with perfect eyes, and degenerating retinas.
So maybe some absolutely miniscule fraction of the poplulation can have blurry monochrome 60x60 pixel vision restored to them. Hopeless for reading... it might be good for navigating sidewalks with a cane.
I'm not arguing that Word is inferior to Wordperfect, only that the whole lump of thick GUI wordprocessors are deceptively intuative -- Word moreso than Wordperfect. It is purely subjective as to whether or not that is an advantage.
I would rather that we had ultra-simple wordprocessors, stripped throughly, people could then send their docs off to the local Word or Wordperfect expert in their group.
When I mentioned bloat it was in reference to data bloat from OLE. The problem I encounter in the field is end-users who cut and paste, only to have objects inserted in their documents... All they really want is CGMs. Most are happy after I explain to them what has happened to their document, and the difference between embedded objects and pasted graphics.
IMHO you are absolutely correct that you cannot pick particular features to remove from a wordprocessor in order to simplify things. But not everybody needs to have a wordprocessor.
I suppose my ideal would be a complex viewer alongside a simple wordprocessor. People could collaborate by sending people gently formatted rich text and in-line graphics (without layout information). For the mostpart simple rich text environments such as Lotus Notes are perfect for this sort of thing. People don't have to worry about page formatting, or even files.
The end result would hopefully be a workforce which respects that Word and Wordperfect are complicated programs and should not be used for hundred-page customer proposals by people who just "figure it out as they go". To stay on topic, they might even respect that erasing super-confidential information from a document might not be as intuative as it seems.
All the while Word uses a cryptic, closed and proprietary document format. That is a huge negative. In the real world, I've seen hundred-page documents crash and burn because somebody tried to embed an object. Funny, Word was able to save the document, it's a shame it wasn't able to load it again.
Other things to consider is that the vast majority of end users only use a minority of features. Most of the people in the real world whom I deal with barely know a tab from an indent from a clump of spaces. They're confused when they change printers and their documents reformat on them.
I would so very much rather give them something like Abiword or even Wordpad and let them write their docs and prepare their graphics, then have somebody who knows what they're doing sit down with the person for an hour and slam everything together.
I've seen sooo much wasted time from lousy features which are half-implemented.
Although Windows and Office have been getting better, I still have to tell people "The reason your document is 34MB, is because you were using cut and paste... that embeds objects by default. this is unreliable, unintuative, unstable and bloated. It's been that way since the feature was first implemented."
So what can Word and all other ridiculous wordprocessors do better than a feature stripped rich text editor?
For the computer literate, they let you mailmerge, save time with styles, create labels, import complex charts while producing professional looking results...
The computer illiterate however, can waste hours on unfriendly though deceptively intuative interfaces. It can also get people killed as critical information is leaked when people think something is deleted. Microsoft and the rest of the industry's philosophy towards user friendliness is inherently flawed. Precision first, user friendliness second... let people read the manual or be fired for their incompetance.
WordPerfect was great... people just had to pick up a manual. And please, don't compare Word 2000 to WordPerfect which hasn't significantly improved since version 6.0 Only recently has Word advanced to the point that I feel a little comfortable that I can trust the software to format my documents without relying upon reveal codes.
Don't kid yourself, the only reason Word is so successful is because of marketing and proprietary formats.
Java, on the other hand, automatically denies any script, or any code downloaded from the network, the privileges to do anything remotely dangerous. If the applet or other piece of code requests permission to do so, the user is given a clear warning that it is dangerous to permit it...
This is a problem waiting to happen. So users click on their friend's email attachment. The user is prompted
The certificate signing authority/Arbitrary wishes permission to the following action: Permission to execute
The end-user, knowing that they want to launch it, think this is silly, and just click "Permit"
Then the user is asked from some obscure signer for permission to access the file system, to access other programs, etcetera. A large enough number of users will think this a nuicense, and just click "Permit" until their attachment runs.
There has to be a better way. If perhaps we could pre-approve all local signing authorities, and refuse everybody's ability to "Permit"... but then one user who knows a little bit too much could spread a virus/trojan/worm through a cooporation like wildfire.
Maybe we should just give up, go back to the CLI and hand all our users manuals. It keeps the stupid people away.
He said double the largest known prime and add two. That way the result is garaunteed not to be a prime (it's even!) and it is not the sum of two primes.
wrong, its not going to be the sum of two known primes.
Oh that's just splitting hairs... I actually meant "the" two primes. You're absolutely correct, but I think it was quite clear that I didn't think it a solution for the problem when I said that all you had to do was calculate which two primes sum to that total.
What may also apply is that the primes may be positive and negative. I don't know.
All primes are positive by definition.
When my statement is qualified with the words "I don't know", you certainly can't fault me for claiming to be an authority on the matter:-)
He said double the largest known prime and add two. That way the result is gauranteed not to be a prime (it's even!) and it is not the sum of two primes.
This and the original post said the sum of two primes. Otherwise we could just keep adding one to our heart's content.
What may also apply is that the primes may be positive and negative. I don't know.
Now it's just a matter of finding two prime factors which sum to a number twice the largest known prime plus two:-)
Slashdot Mirror
Just remember how little you can squeeze in there. Your program and data have to fit in 1024 bytes... unless you get the memory pack.
A PC 80x25 character screen is 2000 characters.
I bought one of these at a garage sale for $15CDN when I was in gradeschool. It was so cool.
I think that exhausts all I have to say about the machine.
You're quite right, it is not easily measured, but it is widely accepted that security holes are often discovered through the act of careless exploits.
It is infinately more difficult to reverse engineer a product than it is to look at the source and study it for weaknesses. At the very least, the source code acts as a guide to explore potential vulnerabilities.
While both IIS and Apache provide people with ample kudos for finding security holes, the attitudes are different. You can't even own a copy of IIS without shelling out for NT server, and then when you do, reverse-engineering puts you in violation of your license agreement. If you were to approach MS with a hole, and somehow convince them that it is a serious issue, you'll be lucky if you're not arrested. If not for piracy, for violation of your License... or you could report it, just give MS a short time to act on the bug, exploit it, make a name for yourself in the news and maybe let a few tools slip.
Hidden developers, lack of source, and potential legal consequences are all disincentive. The only reason to do them the favour when you just spent weeks hacking through a bug, is in fear of their applications failing.
Apache is so much easier. Just post the bug to the developers and be laughed at or be thanked. It's like debugging code written by your own company.
Finding the hole is nowhere near as easy as exploiting it. Not having the source is a major inhibitor to studying the security of an application. Reverse-engineering bugs is a pain in the butt...
How many black-hats knew of the security hole before the one white-hat found it?
How many more security holes are there in the OS/Webserver which we don't know about?
What incentive does Microsoft provide for people to investigate the holes? They don't even provide the source.
Open source projects at least don't inhibit people from finding security holes.
Oh wait... inhibiting somebody from finding a security hole might be part of the NT security model.
Does anybody see an IM application using Gnutella like protocols? Register yourself with your friends and maybe one of thousands of semi-static servers and totally negate the need for AOL/Netscape/Mirabilis and... who's left... Microsoft and Yahoo/Geocities?
Microsoft should change their name to something more catchy... like "Inprise" or "Rebel."
How old was Sunsite when they changed the name?
I get tired of people crying hypocracy when an open forum appears to draw hippcritial conclusions. These are not the opinions of individuals. They are the opinions of those within a group who cry the loudest.
Some people scream about how much the RIAA sucks and how copyrighted music should be free, others scream about the GPL and how it is being violated. There is no reason to believe that these are the same people.
Personally I love the GPL, it lets me write derivative software at work without losing my soul. I hate the RIAA, because I can no longer bear to listen to the radio, television or CDs.
Honour copyrights, not the people who use them to exploit others.
Pirating bad music taken from exploited musicians doesn't help anybody.
This is a very simple idea which hit me when I was walking down the street one day.
All the crappy musicians we here about these days are only heard about because the record industry promotes them. If there were an alternative channel to promote music, maybe I could begin to find some music which doesn't make me wretch. I mean I don't have cable, and I can't bear to listen to the radio... all the loud carpet-shop commercials, "zoom-zoom-zoom" tv-like ads, repeated songs and manipulated news... it just sucks.
What if somebody were to write a magazine which reviews small musicians, and sponsors them to produce MP3s. With a large enough startup capital, they could pay for producers and the like.
The magazine could have two sections... one for musicians which showed remarkable talent and were sponsored. Another section for musicians who are willing to pay to promote themselves. Self-promoting musicians would be clearly indicated... a condition of this could be that the musician has an MP3 file available, and that the musician themselves own their own name and music.
So you sit back, read your magazine, download the MP3s of some bands you've read about etcetera. Maybe it could form a niche. Maybe you don't need radio stations and record labels to popularize music... you could use night clubs, live bands, print and MP3s
This way, the slew of crappy MP3s are filtered out, people have to read to get a sense of which musicians are worthwhile... musicians get some publicity and could potentially earn cash for selling downloads of MP3s...
You could scrap the meatspace printed matter when flat panels become as durable, cheap, high resolution, low power consumption and establish as wide a circulation.
Then it occurred to me that if I've thought of it, somebody else is probably already doing it.
Where can I get this magazine?
DeCSS wouldn't have come about so easily if the author didn't study the encryption mechanisms in Xing's player. That's not nothing.
Otherwise, I agree. It sounds like the ability to destroy another company for using some fundamental concept (such as scrolling left to right, using cascading menus, shooting falling objects from a stationary target, editing text with a keyboard, arrow keys and pgup/pgdn and soforth) is directly proportional to the size of their legal department.
These Hasbro clones on the other hand used the Hasbro games and the strong similarities of the name of the Hasbro games to market their products. If Tetris were called "Blocks" I could understand the re-use. Microsoft certainly couldn't sue for people using the word "Word" in the name of their wordprocessor (which is MS Word compatible, and edits text...with a keyboard, menus, dictionary, etcetera).
There has to be a line somewhere... There is a company which owns the use of the word "Ethical" in the context of Mutual funds. Whenever a company creates an "Ethical" fund and calls it "Ethical" the legal department is engaged. On the other hand, their own funds need not necessarily be "Ethical".
I suppose my point is that it is just as dumb and confused outside the computer industry as it is inside.
Yes, I'm babbling.
Some psychology guy proposed this decades ago... The Whorfian hypothesis. It is a neat idea, although I don't recall why, it was discounted.
A google search reveals oodles of material.
I think 1984 was written around the time this was a big idea.
The "What did you do wrong?" attitude is the same defense used by people who claim that they've been running Win'95 for months without rebooting.
I was installing off a Sony CDU31a on an RLL HDD. That's an XT drive with a proprietary CDROM. Digging up the correct kernel, finding boot parameters, and trying to figure out why I was getting CDR101 read errors while installing took a month.
Come to think of it, I never did figure out the read errors. The drive worked flawlessly under DOS, Win3.1 and OS/2. I wound up copying packages to my FAT partition to install the base packages. Any time I had to read more than a few megs, the CDROM would throw me errors. Despite all this, it was great for installing individual packages after the base was installed.
With an IDE HDD and IDE CDROM, it might have taken me 5 hours...
I agree totally here.
It took me roughly a month to install Slackware back in 1995. I didn't have the internet to help me, and I had some pretty unusual hardware, but DOS and Windows didn't have any trouble at all.
Windows 3.1 was by far more memory efficient and faster. X was unusable on an unaccellerated 640x480 ISA VGA adapter... unlike Win3.1.
I did some Fortran programming under the environment, which I found useful... but it was easier to dial into the VAX. The fonts were terrible, printer support was almost nonexistant, so for me it was effectively a bloated programming environment.
I administered a small dial up ISP for a short period in 1997. They were running NT3.51, and were scrambling to try to put together a decent system for software development... I put the whole operation under Linux. No big deal. Everything was completely free and worked flawlessly.
Since 1995, I've tried many times to establish a reasonable user environment under Linux. I've tried Gnome and KDE, and they haven't done anything other than to promise me what I had on a 386 running Win3.1... that is to say a reasonable printing architecture, some decent fonts, some standard keystrokes and cut-and-paste. I've tried simplifying things... for some time I ran ICEWM (which I really do like) and then proceeded to try to set up an email client.
There isn't anything but promises here too. Netscape appears to be the best choice... unless I go back to PINE. So I tried to configure Netscape... it crashes. I try some more, it crashes again... infact it crashes quite reliably under many different scenarios... none of which are preventable. So I give up. I tried Spruce, which though promising, was simply incomplete.
I then try to hear multiple audio streams simultaneously. After installing ALSA, reading many FAQs and getting it running, I learn that the Advanced Linux Sound Architecture doesn't support multiple simultaneous streams... like Win3.1 did. So I installed ESD, then I installed Real Player. After a week of applying various binary patches, reinstalling, reading reams of documentation describing how Real Inc. depended upon undocumented features and broke in V5, I finally got some alpha version of the player working and started listening to some Internet radio broadcasts.
So I began working on tracking down another email client. Inches from headed back to PINE, reading up on fetchmail and procmail, Real Player freezes.
Many other experiments revealed that it always would do this some 15 minutes into the broadcast.
On another forray into the Linux world, I installed Gnomehack... it worked fine for a while, and I reminisced about playing hack for long hours ... When I went to restore a game one day, X crashed.
It was remarkable. I launch the game, Gnomehack and X crashes, taking down all my other applications, loosing all my data, reliably, reproducably, and inexplicably. It obviously is a bug in my X server.
After many forrays, and certainly many more to come, I have come to the conclusion that the best server OS for a small company or even for a home user is Linux. It is stable, not price prohibative, will teach you a lot about how computers work, and has many other bennefits. The best GUI for that operating system is either Windows, MacOS, OS/2 or just about anything else you can think of. Linux handles itself so well in a networked environment that there is no reason to put a keyboard or monitor on the device.
I can use Pine just as well on my remote linux box as I can locally, and I'll even be able to fill the fonts crisply out to the edge of the screen, and even cut and paste from it.... while listening to radio broadcasts, and chatting on that whore of a program called ICQ... and I will spend less time reinstalling the operating system for the life of the hardware than I will configuring Linux to do the same.
To Linux's credit, Linux is closer to a good UI than Microsoft is to a robust multiuser operating system.
Apply for a teaching position at a university? Maybe they'll print your book... yes, they may charge in excess of $200 for it, but you'll be paid to teach the material you're writing about.
Personally I don't see the value of printing the book. I imagine O'Reilly isn't interested because it wasn't a good seller. This is no reflection on the quality of the book (I've skimmed it and put it back on the shelf), just a reflection on the size of the market.
(For those going to respond that Linux is huge, yadda yadda yadda, I mean a small market of Linux developers which work on TCP/IP programs, and have found the existing code, documentation and community support insufficient.)
The market of people willing to sign up for a detailed course on Network protocols, at which the core material is Linux's TCP/IP stack might be much more lucrative.
For all you know he might be writing on a machine which doesn't have lowercase characters... like... well... god, I have no idea what you would use to connect from your office/whatever in those days...
A Timex Sinclair 1000 desparately soldered togther to emulate an RS232 port?
Unless there were actually VTs which were case impaired....
Maybe a PET which would require switching video modes to see 40 columns of lower case characters.
Because it won't be a CD, it will be an audio-DVD. Also, the DMCA makes any attempts to circumvent the system illegal -- even going down to Radio Shack and buying an Op-amp so that you can make a backup copy of the media or use soundbites from the disk for your default beep.
Yup... anybody who's handled liquid Nitrogen in a confined space knows... displacing vast quantities of oxygen is a bad thing.
What about the Thinkpad 240 or 570?
They're not the best for portability or battery life, but they would be perfect candidates for what the Cursoe chip claims to be. Have you ever seen a 240?
Whoops... Correction. The old one required external prosthetics.
The technology was featured at least twice before on Slashdot, and I saw it in a documentry over a year ago.
Here's one of them, I can't find the better ones:
I have to wonder if it will even help with retina detachments, the nerve endings wouldn't be there to interface with... it seems to be limited to helping people with perfect eyes, and degenerating retinas.
So maybe some absolutely miniscule fraction of the poplulation can have blurry monochrome 60x60 pixel vision restored to them. Hopeless for reading... it might be good for navigating sidewalks with a cane.
The real beauty of Minidiscs is the ability to record near perfect digital audio from a relatively inexpensive and very portable device.
Nothing on the market challenges this niche yet.
I'm not arguing that Word is inferior to Wordperfect, only that the whole lump of thick GUI wordprocessors are deceptively intuative -- Word moreso than Wordperfect. It is purely subjective as to whether or not that is an advantage.
I would rather that we had ultra-simple wordprocessors, stripped throughly, people could then send their docs off to the local Word or Wordperfect expert in their group.
When I mentioned bloat it was in reference to data bloat from OLE. The problem I encounter in the field is end-users who cut and paste, only to have objects inserted in their documents... All they really want is CGMs. Most are happy after I explain to them what has happened to their document, and the difference between embedded objects and pasted graphics.
IMHO you are absolutely correct that you cannot pick particular features to remove from a wordprocessor in order to simplify things. But not everybody needs to have a wordprocessor.
I suppose my ideal would be a complex viewer alongside a simple wordprocessor. People could collaborate by sending people gently formatted rich text and in-line graphics (without layout information). For the mostpart simple rich text environments such as Lotus Notes are perfect for this sort of thing. People don't have to worry about page formatting, or even files.
The end result would hopefully be a workforce which respects that Word and Wordperfect are complicated programs and should not be used for hundred-page customer proposals by people who just "figure it out as they go". To stay on topic, they might even respect that erasing super-confidential information from a document might not be as intuative as it seems.
All the while Word uses a cryptic, closed and proprietary document format. That is a huge negative. In the real world, I've seen hundred-page documents crash and burn because somebody tried to embed an object. Funny, Word was able to save the document, it's a shame it wasn't able to load it again.
Other things to consider is that the vast majority of end users only use a minority of features. Most of the people in the real world whom I deal with barely know a tab from an indent from a clump of spaces. They're confused when they change printers and their documents reformat on them.
I would so very much rather give them something like Abiword or even Wordpad and let them write their docs and prepare their graphics, then have somebody who knows what they're doing sit down with the person for an hour and slam everything together.
I've seen sooo much wasted time from lousy features which are half-implemented.
Although Windows and Office have been getting better, I still have to tell people "The reason your document is 34MB, is because you were using cut and paste... that embeds objects by default. this is unreliable, unintuative, unstable and bloated. It's been that way since the feature was first implemented."
So what can Word and all other ridiculous wordprocessors do better than a feature stripped rich text editor?
For the computer literate, they let you mailmerge, save time with styles, create labels, import complex charts while producing professional looking results...
The computer illiterate however, can waste hours on unfriendly though deceptively intuative interfaces. It can also get people killed as critical information is leaked when people think something is deleted. Microsoft and the rest of the industry's philosophy towards user friendliness is inherently flawed. Precision first, user friendliness second... let people read the manual or be fired for their incompetance.
WordPerfect was great... people just had to pick up a manual. And please, don't compare Word 2000 to WordPerfect which hasn't significantly improved since version 6.0 Only recently has Word advanced to the point that I feel a little comfortable that I can trust the software to format my documents without relying upon reveal codes.
Don't kid yourself, the only reason Word is so successful is because of marketing and proprietary formats.
Java, on the other hand, automatically denies any script, or any code downloaded from the network, the privileges to do anything remotely dangerous. If the applet or other piece of code requests permission to do so, the user is given a clear warning that it is dangerous to permit it...
This is a problem waiting to happen. So users click on their friend's email attachment. The user is prompted
The end-user, knowing that they want to launch it, think this is silly, and just click "Permit"
Then the user is asked from some obscure signer for permission to access the file system, to access other programs, etcetera. A large enough number of users will think this a nuicense, and just click "Permit" until their attachment runs.
There has to be a better way. If perhaps we could pre-approve all local signing authorities, and refuse everybody's ability to "Permit"... but then one user who knows a little bit too much could spread a virus/trojan/worm through a cooporation like wildfire.
Maybe we should just give up, go back to the CLI and hand all our users manuals. It keeps the stupid people away.
He said double the largest known prime and add two. That way the result is garaunteed not to be a prime (it's even!) and it is not the sum of two primes.
wrong, its not going to be the sum of two known primes.
Oh that's just splitting hairs... I actually meant "the" two primes. You're absolutely correct, but I think it was quite clear that I didn't think it a solution for the problem when I said that all you had to do was calculate which two primes sum to that total.
What may also apply is that the primes may be positive and negative. I don't know.
All primes are positive by definition.
When my statement is qualified with the words "I don't know", you certainly can't fault me for claiming to be an authority on the matter :-)
But thanks for the correction.
He said double the largest known prime and add two. That way the result is gauranteed not to be a prime (it's even!) and it is not the sum of two primes.
This and the original post said the sum of two primes. Otherwise we could just keep adding one to our heart's content.
What may also apply is that the primes may be positive and negative. I don't know.
Now it's just a matter of finding two prime factors which sum to a number twice the largest known prime plus two :-)
That sounds like a blatent ripoff of Pimp Wars.