Entertainment has its own trademark category. IANAL but unless Toho is offering Godzilla-branded telecommunication services, there's no likelihood of customer confusion and Toho has no legitimate right to crack down on unrelated uses (they're just proving the system can be abused by anyone with deep pockets).
Trademarks aren't about originality but authenticity and reputation. "Stealing" ideas is good--that's how civilization advances!
A short story called "Flatlander". It's collected in Crashlander (rather than Flatlander) because it's about Beowulf Shaeffer, along with a rich Earth native who goes by "Elephant" gaining a healthy sense of fear.
I don't want to sift random cruft out of a collection. I want to declare the type of the contents and eliminate type errors at compile time. That's what statically-typed languages are for. There are plenty of functional and prototype-based languages (many targeting JVM bytecode) I'd use if I wanted type checks postponed for runtime.
Re:sprintf is the wrong analogy
on
Shattering Windows
·
· Score: 2, Insightful
Only a similar problem. Unless there's something dire about xlib that I've forgotten, POSIX systems don't tend to bundle user interface libraries that accept messages from any user, interpret a field as a function pointer in the server's address space, and immediately call that function whenever the app doesn't specifically overrides that behavior. That's somewhat exploitable even if you don't have a buffer overrun to take advantage of.
He's not on drugs re int and BigInteger. It doesn't take much runtime support to shift between bignums (boxed) and fixnums (stored in most of the bits of a pointer, with a few left over as a tag). Lisp vendors have offered this for decades, complete with optimization (store a native integer if a value is proved or declared to stay in its range).
There's always LS-120, which uses 3.5" diskettes with a laser etched servo control pattern to reach 120MB worth of density.
It never caught on because Zip was cheaper and the public didn't hear about the Click of Death until too late.
The only way you could possibly edit a file without reading and writing all the way to the end is to seek to somewhere in the middle while making sure you modify bytes without inserting or deleting any. Is that really useful?
Show us another cross-platform editor that can drive a version control system, read and write multibyte character encodings, and jump to source from error messages out of most any compiler or SGML validator and I'll consider it. Helps if it works over a tty and if third parties can integrate a PIM and newsreader.
I'm quite sure that if I did something illegal on my own time, my corporate masters wouldn't get dragged into court for it.
Even the threat of it could be unacceptable if your pockets are deep enough, and the US legal system is screwed up enough that I can't even say you wouldn't lose.
Most companies want their staff to consist of the cheapest employees who can just uphold existing commitments (those it's not cheaper to break) until their markets improve. There are few entrepreneuers out there with the money and courage to start new offerings, which is a shame considering the bargains they'd have on resources.
Sorry, I was thinking of those old xmodmap files that swap the two, leaving you with just one Ctrl key on the left and putting you through risky contortions to touch-type Ctrl-Q Ctrl-R, for example.
Unless you have another Caps Lock key on the right side, swapping with Ctrl is not a good idea. Holding down Ctrl/Shift/Meta and striking another key with the same hand is not only slower (once you've learned to touch-type correctly) but allegedly bad for your wrists.
Early merging causes larger traffic jams. You're effectively lengthening the lane closure, leaving less pavement for the same cars to use (traffic goes slower when you compress it) and having dozens of drivers jockeying for position over thousands of feet rather than taking turns at the one obvious place. If the engineers wanted you off that strip of pavement, they would have moved the lane closure further ahead.
How can you possibly get "much simpler" than this?
#define property(name, type) \ type name##_; \ type name() const { return name##_; } \ type name(type new_value) { \ type old_value(name##_); \ name##_ = new_value; \ return old_value; \ }
RTTI for members is a good idea, but altering the language to enable a feature that doesn't even change the meaning of the code is ridiculous. Why not just provide it for all members with a switch or #pragma for the rare cases when you really can't afford it?
To be pedantic, mapping to colors from a smaller palette is is quantization. Dithering is injecting noise during quantization in an attempt to reduce overall error. You can also quantize without dithering, which is sometimes called posterizing.
then [the infringed party] are restricted from receiving royalties that accrue prior to the notice of infringement.
It's important to note that even after laches, they can sue for all infringement after notification, and so the technology still becomes unavailable to our generation.
If they were too busy on other projects, had retired or died, or demanded an unreasonable rate, wouldn't a venue for getting bids from other qualified maintainers have been useful?
That gives the original developer a near-monopoly on making changes. A task market that isn't controlled by any project would allow competition and help establish fairer prices in for-hire maintenance.
... as in speech, of course (I call free-as-in-beer "cheap"). Free software would be better off if more people commissioned development. At the moment the features we have are mostly limited to those so fascinating to create (or maybe valuable on a rsum) that the developers donated their labor. Even better if we can coordinate many customers' milder desires so that none of them need bear the entire cost--the result would be funded much like commercial software today, but as a sort of work for hire for the community it wouldn't be controlled by the original author(s).
An ad hoc parser isn't going to correctly handle character entity references, comments, marked sections, processing instructions, and non-UTF-8 encodings, and all of these are permitted for application/beep+xml (everything but DOCTYPE declarations, which makes the DTD they wrote pretty pointless).
Every BEEP-aware system must incorporate a complete XML parser and MIME header tokenizer, and fire them up every time a session or channel opens or closes. Since just about everything in a BEEP frame is variable length and channel numbers are arbitrary, you can't prioritize, scan, or filter any channel without vastly more horsepower than anyone's routers or firewalls have.
What I really can't figure out is why the first (and only) mapping document is for TCP. BEEP doesn't even buy you anything unless you can retransmit frames out of order.
Without Carmack, commodity video hardware wouldn't support OpenGL, so serious 3D modelling packages wouldn't be available on the desktop. You'd still be in the bad old days of paying a mint for "engineering workstations".
You can lease without much of any down payment, and the payments are lower (they don't have to account for the entire value of the car) and easily charged against business income for taxes (if applicable). Buying makes sense if you're going to put heavy wear on the car (which leases penalize), or your mechanic will help you eke every last mile out of it with accompanying downtime. Leasing is more like a long-term rental of a car you don't intend to keep permanently (maybe a fancier one than you could really buy) on much less onerous terms than these nanny-state sleazeballs demand.
Slashdot Mirror
Entertainment has its own trademark category. IANAL but unless Toho is offering Godzilla-branded telecommunication services, there's no likelihood of customer confusion and Toho has no legitimate right to crack down on unrelated uses (they're just proving the system can be abused by anyone with deep pockets).
Trademarks aren't about originality but authenticity and reputation. "Stealing" ideas is good--that's how civilization advances!
A short story called "Flatlander". It's collected in Crashlander (rather than Flatlander) because it's about Beowulf Shaeffer, along with a rich Earth native who goes by "Elephant" gaining a healthy sense of fear.
I don't want to sift random cruft out of a collection. I want to declare the type of the contents and eliminate type errors at compile time. That's what statically-typed languages are for. There are plenty of functional and prototype-based languages (many targeting JVM bytecode) I'd use if I wanted type checks postponed for runtime.
Only a similar problem. Unless there's something dire about xlib that I've forgotten, POSIX systems don't tend to bundle user interface libraries that accept messages from any user, interpret a field as a function pointer in the server's address space, and immediately call that function whenever the app doesn't specifically overrides that behavior. That's somewhat exploitable even if you don't have a buffer overrun to take advantage of.
He's not on drugs re int and BigInteger. It doesn't take much runtime support to shift between bignums (boxed) and fixnums (stored in most of the bits of a pointer, with a few left over as a tag). Lisp vendors have offered this for decades, complete with optimization (store a native integer if a value is proved or declared to stay in its range).
There's always LS-120, which uses 3.5" diskettes with a laser etched servo control pattern to reach 120MB worth of density. It never caught on because Zip was cheaper and the public didn't hear about the Click of Death until too late.
The only way you could possibly edit a file without reading and writing all the way to the end is to seek to somewhere in the middle while making sure you modify bytes without inserting or deleting any. Is that really useful?
Show us another cross-platform editor that can drive a version control system, read and write multibyte character encodings, and jump to source from error messages out of most any compiler or SGML validator and I'll consider it. Helps if it works over a tty and if third parties can integrate a PIM and newsreader.
Even the threat of it could be unacceptable if your pockets are deep enough, and the US legal system is screwed up enough that I can't even say you wouldn't lose.
Most companies want their staff to consist of the cheapest employees who can just uphold existing commitments (those it's not cheaper to break) until their markets improve. There are few entrepreneuers out there with the money and courage to start new offerings, which is a shame considering the bargains they'd have on resources.
Sorry, I was thinking of those old xmodmap files that swap the two, leaving you with just one Ctrl key on the left and putting you through risky contortions to touch-type Ctrl-Q Ctrl-R, for example.
Unless you have another Caps Lock key on the right side, swapping with Ctrl is not a good idea. Holding down Ctrl/Shift/Meta and striking another key with the same hand is not only slower (once you've learned to touch-type correctly) but allegedly bad for your wrists.
Early merging causes larger traffic jams. You're effectively lengthening the lane closure, leaving less pavement for the same cars to use (traffic goes slower when you compress it) and having dozens of drivers jockeying for position over thousands of feet rather than taking turns at the one obvious place. If the engineers wanted you off that strip of pavement, they would have moved the lane closure further ahead.
How can you possibly get "much simpler" than this?
RTTI for members is a good idea, but altering the language to enable a feature that doesn't even change the meaning of the code is ridiculous. Why not just provide it for all members with a switch or #pragma for the rare cases when you really can't afford it?
To be pedantic, mapping to colors from a smaller palette is is quantization. Dithering is injecting noise during quantization in an attempt to reduce overall error. You can also quantize without dithering, which is sometimes called posterizing.
If the patent is invalid, that means it always was, so why can't licensees sue for fraud?
It's important to note that even after laches, they can sue for all infringement after notification, and so the technology still becomes unavailable to our generation.
If they were too busy on other projects, had retired or died, or demanded an unreasonable rate, wouldn't a venue for getting bids from other qualified maintainers have been useful?
That gives the original developer a near-monopoly on making changes. A task market that isn't controlled by any project would allow competition and help establish fairer prices in for-hire maintenance.
... as in speech, of course (I call free-as-in-beer "cheap"). Free software would be better off if more people commissioned development. At the moment the features we have are mostly limited to those so fascinating to create (or maybe valuable on a rsum) that the developers donated their labor. Even better if we can coordinate many customers' milder desires so that none of them need bear the entire cost--the result would be funded much like commercial software today, but as a sort of work for hire for the community it wouldn't be controlled by the original author(s).
An ad hoc parser isn't going to correctly handle character entity references, comments, marked sections, processing instructions, and non-UTF-8 encodings, and all of these are permitted for application/beep+xml (everything but DOCTYPE declarations, which makes the DTD they wrote pretty pointless).
What I really can't figure out is why the first (and only) mapping document is for TCP. BEEP doesn't even buy you anything unless you can retransmit frames out of order.
Without Carmack, commodity video hardware wouldn't support OpenGL, so serious 3D modelling packages wouldn't be available on the desktop. You'd still be in the bad old days of paying a mint for "engineering workstations".
Didn't Microsoft pay ActiveState to port Perl to Win32 in the first place, just so a few more people would take IIS seriously?
You can lease without much of any down payment, and the payments are lower (they don't have to account for the entire value of the car) and easily charged against business income for taxes (if applicable). Buying makes sense if you're going to put heavy wear on the car (which leases penalize), or your mechanic will help you eke every last mile out of it with accompanying downtime. Leasing is more like a long-term rental of a car you don't intend to keep permanently (maybe a fancier one than you could really buy) on much less onerous terms than these nanny-state sleazeballs demand.